WO2013129352A1

WO2013129352A1 - Three-dimensional object detection device

Info

Publication number: WO2013129352A1
Application number: PCT/JP2013/054853
Authority: WO
Inventors: 早川　泰久; 修深田
Original assignee: 日産自動車株式会社
Priority date: 2012-03-01
Filing date: 2013-02-26
Publication date: 2013-09-06
Also published as: BR112014020406B1; RU2633120C2; MY173364A; US9830519B2; US20150002671A1; EP2821981A4; CN104115204A; BR112014020406A2; CN104115204B; MX2014009700A; JPWO2013129352A1; RU2014139679A; JP5743020B2; EP2821981A1; MX344875B

Abstract

　The present invention is provided with: an imaging means (10) that images a predetermined area rearward of a host vehicle; an image conversion means (31) that converts the viewpoint of the captured image to an overhead view image; a three-dimensional object detection means (32, 33) that generates difference waveform information by counting and creating a frequency distribution of the number of pixels indicating a difference equal to or greater than a predetermined first threshold value in a difference image in which overhead view images taken at different times are aligned, and, if the difference waveform information is equal to or greater than a predetermined second threshold value, detects a three-dimensional object present in an adjacent lane on the basis of the difference waveform information; and a light source detection means (34) that detects a light source present rearward of the host vehicle. If the light source detection means (34) detects a light source, the first and second threshold values are set higher in a forward area farther ahead than a line that connects the light source and the imaging means (10) in the predetermined area, such that three-dimensional objects are more difficult to detect.

Description

Three-dimensional object detection device

The present invention relates to a three-dimensional object detection device.
This application claims priority based on Japanese Patent Application No. 2012-045349 filed on Mar. 1, 2012. For designated countries that are allowed to be incorporated by reference, The contents described in the application are incorporated into the present application by reference and made a part of the description of the present application.

2. Description of the Related Art Conventionally, a technique is known in which two captured images captured at different times are converted into a bird's-eye view image, and an obstacle is detected based on the difference between the two converted bird's-eye view images (Patent Literature). 1).

JP 2008-227646 A

When detecting other vehicles traveling in an adjacent lane adjacent to the lane in which the host vehicle travels using an image obtained by capturing the rear of the host vehicle at night, the lane adjacent to the adjacent lane (hereinafter also referred to as the adjacent adjacent lane) When the light of the headlight of another vehicle that travels is irradiated on the detection area, the headlight light image may be erroneously detected as another vehicle traveling in the adjacent lane. Headlight light is characterized by high brightness, so it is possible to identify the light of the headlight of a vehicle traveling in the adjacent lane according to the brightness, but the lens of the imaging device is dirty The problem is that the light of the headlights of other vehicles traveling in the adjacent adjacent lane is diffusely reflected, and the brightness around the headlights increases, so that the adjacent vehicles located around the headlights cannot be detected properly. was there.

The problem to be solved by the present invention is to eliminate the influence of the headlight of another vehicle and detect the other vehicle traveling in the adjacent lane when detecting the other vehicle traveling in the adjacent lane adjacent to the lane in which the host vehicle is traveling. Is accurately detected.

The present invention detects a light source existing behind the host vehicle and detects a three-dimensional object existing in an adjacent lane so that a three-dimensional object is difficult to detect in a region ahead of the line connecting the detected light source and the imaging device. The above-mentioned problem is solved by increasing the detection standard for doing so.

According to the present invention, in the area in front of the headlight (light source), by increasing the detection standard of the adjacent vehicle, it is possible to effectively prevent erroneous detection of the adjacent vehicle around the headlight, and from the headlight (light source). Since the tire wheel of the adjacent lane existing behind can be detected appropriately, the adjacent vehicle traveling in the adjacent lane can be detected appropriately.

FIG. 1 is a schematic configuration diagram of a vehicle equipped with the three-dimensional object detection device according to the first embodiment. FIG. 2 is a plan view showing a traveling state of the vehicle of FIG. FIG. 3 is a block diagram showing details of the computer according to the first embodiment. FIGS. 4A and 4B are diagrams for explaining the outline of the processing of the alignment unit according to the first embodiment, in which FIG. 4A is a plan view showing the moving state of the vehicle, and FIG. 4B is an image showing the outline of alignment. It is. FIG. 5 is a schematic diagram illustrating how a differential waveform is generated by the three-dimensional object detection unit according to the first embodiment. FIG. 6 is a diagram for explaining a method of detecting an adjacent vehicle according to the first embodiment. Figure 7 is a graph showing the brightness at the detected position of the detection area A1, A2, an example of a first relation between the threshold alpha _1. Figure 8 is a diagram for explaining a rear distance from the camera, the first relationship between the threshold alpha _1. FIG. 9A is a diagram illustrating a scene in which the light of the headlight of the adjacent vehicle is irradiated on the region R2 in the detection regions A1 and A2. FIG. 9B is a diagram illustrating a scene in which the light of the headlight of the adjacent vehicle is irradiated on the region R1 in the detection regions A1 and A2. FIG. 9C is a diagram illustrating a scene in which the light of the headlight of the adjacent vehicle is irradiated on the region R3 in the detection regions A1 and A2. FIG. 10 is a graph showing an example of a first threshold map showing the relationship between the brightness at the detection positions in the detection areas A1 and A2 and the threshold α ₂ ′. FIG. 11 is a diagram showing an example of a second threshold map showing the relationship between the luminance at the detection positions in the detection areas A1 and A2 and the threshold α ₂ ″. FIG. 12 is a diagram illustrating an example of a control map that defines the weight wt of the threshold value α ₂ ″ obtained from the second threshold map. FIG. 13 is a diagram for explaining a method of adjusting the control map shown in FIG. FIG. 14 is a diagram illustrating a small region divided by the three-dimensional object detection unit according to the first embodiment. FIG. 15 is a diagram illustrating an example of a histogram obtained by the three-dimensional object detection unit according to the first embodiment. FIG. 16 is a diagram illustrating weighting by the three-dimensional object detection unit according to the first embodiment. FIG. 17 is a diagram illustrating another example of a histogram obtained by the three-dimensional object detection unit according to the first embodiment. FIG. 18 is a flowchart illustrating the adjacent vehicle detection method according to the first embodiment. FIG. 19 is a flowchart showing the threshold value α setting process in step S105. FIG. 20 is a block diagram illustrating details of the computer according to the second embodiment. FIGS. 21A and 21B are diagrams illustrating the traveling state of the vehicle, in which FIG. 21A is a plan view illustrating the positional relationship of the detection region and the like, and FIG. 21B is a perspective view illustrating the positional relationship of the detection region and the like in real space. FIG. 22 is a diagram for explaining the operation of the luminance difference calculation unit according to the second embodiment. FIG. 22A is a diagram showing the positional relationship among the attention line, reference line, attention point, and reference point in the bird's-eye view image. (B) is a figure which shows the positional relationship of the attention line, reference line, attention point, and reference point in real space. 23A and 23B are diagrams for explaining the detailed operation of the luminance difference calculation unit according to the second embodiment, in which FIG. 23A is a diagram showing a detection area in a bird's-eye view image, and FIG. 23B is an attention in a bird's-eye view image. It is a figure which shows the positional relationship of a line, a reference line, an attention point, and a reference point. FIG. 24 is a diagram illustrating an example of an image for explaining the edge detection operation. Figure 25 is a graph showing the brightness at the detected position of the detection area A1, A2, an example of a third relationship between the threshold value beta _1. Figure 26 is a diagram for explaining a rear distance from the camera, a third relationship between the threshold value beta _1. FIG. 27 is a diagram illustrating an example of a third threshold map indicating the relationship between the luminance in the detection areas A1 and A2 and the threshold β ₂ ′. FIG. 28 is a diagram showing an example of a fourth threshold map showing the relationship between the luminance in the detection areas A1 and A2 and the threshold value β ₂ ″. FIG. 29 is a diagram illustrating an example of a control map defining the weight wt of the threshold value β ₂ ″ obtained from the fourth threshold map. 30A and 30B are diagrams showing the edge line and the luminance distribution on the edge line, where FIG. 30A is a diagram showing the luminance distribution when a three-dimensional object (adjacent vehicle) is present in the detection area, and FIG. 30B is the detection area. It is a figure which shows luminance distribution when a solid object does not exist in FIG. FIG. 31 is a flowchart showing an adjacent vehicle detection method according to the second embodiment. FIG. 32 is a diagram illustrating a control map according to another embodiment.

<< First Embodiment >>
FIG. 1 is a schematic configuration diagram of a vehicle equipped with a three-dimensional object detection device 1 according to the first embodiment. The three-dimensional object detection device 1 according to the present embodiment is intended to detect a vehicle (hereinafter also referred to as an adjacent vehicle V2) that exists in an adjacent lane that may be contacted when the host vehicle V1 changes lanes. To do. As shown in FIG. 1, the three-dimensional object detection device 1 according to the present embodiment includes a camera 10, a vehicle speed sensor 20, and a calculator 30.

As shown in FIG. 1, the camera 10 is attached to the vehicle V <b> 1 so that the optical axis is at an angle θ downward from the horizontal at a position of the height h behind the host vehicle V <b> 1. The camera 10 captures an image of a predetermined area in the surrounding environment of the host vehicle V1 from this position. The vehicle speed sensor 20 detects the traveling speed of the host vehicle V1, and calculates the vehicle speed from the wheel speed detected by, for example, a wheel speed sensor that detects the rotational speed of the wheel. The computer 30 detects an adjacent vehicle existing in an adjacent lane behind the host vehicle.

FIG. 2 is a plan view showing a traveling state of the host vehicle V1 of FIG. As shown in the figure, the camera 10 images the vehicle rear side at a predetermined angle of view a. At this time, the angle of view a of the camera 10 is set to an angle of view at which the left and right lanes (adjacent lanes) can be imaged in addition to the lane in which the host vehicle V1 travels.

FIG. 3 is a block diagram showing details of the computer 30 of FIG. In FIG. 3, the camera 10 and the vehicle speed sensor 20 are also illustrated in order to clarify the connection relationship.

As shown in FIG. 3, the computer 30 includes a viewpoint conversion unit 31, a positioning unit 32, a three-dimensional object detection unit 33, a certainty factor determination unit 34, and a detection reference setting unit 35. Below, each structure is demonstrated.

The viewpoint conversion unit 31 inputs captured image data of a predetermined area obtained by imaging with the camera 10, and converts the viewpoint of the input captured image data into bird's-eye image data in a bird's-eye view state. The state viewed from a bird's-eye view is a state viewed from the viewpoint of a virtual camera looking down from above, for example, vertically downward. This viewpoint conversion can be executed as described in, for example, Japanese Patent Application Laid-Open No. 2008-219063. The viewpoint conversion of captured image data to bird's-eye view image data is based on the principle that a vertical edge peculiar to a three-dimensional object is converted into a straight line group passing through a specific fixed point by viewpoint conversion to bird's-eye view image data. This is because a planar object and a three-dimensional object can be distinguished if used.

The alignment unit 32 sequentially inputs the bird's-eye view image data obtained by the viewpoint conversion of the viewpoint conversion unit 31 and aligns the positions of the inputted bird's-eye view image data at different times. 4A and 4B are diagrams for explaining the outline of the processing of the alignment unit 32, where FIG. 4A is a plan view showing the moving state of the host vehicle V1, and FIG. 4B is an image showing the outline of the alignment.

As shown in FIG. 4 (a), the host vehicle V1 of the current time is located in P _1, one unit time before the vehicle V1 is located in the P _{1 '.} Further, there is a parallel running state with the vehicle V1 is located is adjacent vehicle V2 laterally after the vehicle V1, located in P ₂ adjacent vehicle V2 is the current time, one unit time before the adjacent vehicle V2 is P ₂ Suppose it is located at '. Furthermore, it is assumed that the host vehicle V1 has moved a distance d at one time. Note that “one hour before” may be a past time for a predetermined time (for example, one control cycle) from the current time, or may be a past time for an arbitrary time.

In this state, the bird's-eye view image PB _t at the current time is as shown in Figure 4 (b). In the bird's-eye view image PB _t, becomes a rectangular shape for the white line drawn on the road surface, but a relatively accurate is a plan view state, the adjacent vehicle V2 (position P ₂₎ is tilting occurs. Similarly, in the bird's-eye view image PB _{t-1 one} hour before, the white line drawn on the road surface has a rectangular shape, and is in a state of being relatively accurately viewed in plan, but the adjacent vehicle V2 (position P _2). ') Will fall down. As described above, the vertical edges of solid objects (including the edges that rise in the three-dimensional space from the road surface in addition to the vertical edges in the strict sense) are straight lines along the collapse direction by the viewpoint conversion processing to bird's-eye view image data. This is because the plane image on the road surface does not include a vertical edge, but such a fall does not occur even when the viewpoint is changed.

The alignment unit 32 performs alignment of the bird's-eye view images PB _t and PB _t−1 as described above on the data. At this time, the alignment unit 32 offsets the bird's-eye view image PB _t-1 at the previous time and matches the position with the bird's-eye view image PB _{t at the} current time. The image on the left side and the center image in FIG. 4B show a state that is offset by the movement distance d ′. This offset amount d ′ is a movement amount on the bird's-eye view image data corresponding to the actual movement distance d of the host vehicle V1 shown in FIG. 4 (a). It is determined based on the time until the time.

Further, after the alignment, the alignment unit 32 takes the difference between the bird's-eye view images PB _t and PB _t−1 and generates data of the difference image PD _t . Here, in the present embodiment, the alignment unit 32 converts the pixel value difference between the bird's-eye view images PB _t and PB _t−1 to an absolute value in order to cope with a change in the illumination environment, and the absolute value is a predetermined value. By setting the pixel value of the difference image PD _t to “1” when the threshold value is equal to or greater than the threshold th, and setting the pixel value of the difference image PD _t to “0” when the absolute value is less than the predetermined threshold th, Data of the difference image PD _t as shown on the right side of FIG. 4B can be generated.

Returning to FIG. 3, the three-dimensional object detection unit 33 detects a three-dimensional object based on the data of the difference image PD _t shown in FIG. At this time, the three-dimensional object detection unit 33 also calculates the movement distance of the three-dimensional object in the real space. In detecting the three-dimensional object and calculating the movement distance, the three-dimensional object detection unit 33 first generates a differential waveform.

In generating the difference waveform, the three-dimensional object detection unit 33 sets a detection region in the difference image PD _t . The three-dimensional object detection device 1 of the present example is intended to calculate a movement distance for an adjacent vehicle that may be contacted when the host vehicle V1 changes lanes. For this reason, in this example, as shown in FIG. 2, rectangular detection areas A1, A2 are set on the rear side of the host vehicle V1. Such detection areas A1, A2 may be set from a relative position with respect to the host vehicle V1, or may be set based on the position of the white line. When setting the position of the white line as a reference, the three-dimensional object detection device 1 may use, for example, an existing white line recognition technique.

Further, as shown in FIG. 2, the three-dimensional object detection unit 33 recognizes the sides (sides along the traveling direction) of the set detection areas A1 and A2 on the own vehicle V1 side as the ground lines L1 and L2. In general, the ground line means a line in which the three-dimensional object contacts the ground. However, in the present embodiment, the ground line is set as described above, not a line in contact with the ground. Even in this case, from experience, the difference between the ground line according to the present embodiment and the ground line obtained from the position of the original adjacent vehicle V2 is not too large, and there is no problem in practical use.

FIG. 5 is a schematic diagram illustrating how the three-dimensional object detection unit 33 generates a differential waveform. As shown in FIG. 5, the three-dimensional object detection unit 33 calculates a differential waveform from a portion corresponding to the detection areas A <b> 1 and A <b> 2 in the difference image PD _t (right diagram in FIG. 4B) calculated by the alignment unit 32. DW _t is generated. At this time, the three-dimensional object detection unit 33 generates a differential waveform DW _t along the direction in which the three-dimensional object falls by viewpoint conversion. In the example shown in FIG. 5, only the detection area A1 is described for convenience, but the difference waveform DW _t is generated for the detection area A2 in the same procedure.

Specifically, first three-dimensional object detection unit 33 defines a line La on the direction the three-dimensional object collapses on data of the difference image PD _t. Then, the three-dimensional object detection unit 33 counts the number of difference pixels DP indicating a predetermined difference on the line La. In the present embodiment, the difference pixel DP indicating the predetermined difference is expressed by the pixel value of the difference image PD _t as “0” and “1”, and the pixel indicating “1” is counted as the difference pixel DP. .

The three-dimensional object detection unit 33 counts the number of difference pixels DP and then obtains an intersection point CP between the line La and the ground line L1. Then, the three-dimensional object detection unit 33 associates the intersection CP with the count number, determines the horizontal axis position based on the position of the intersection CP, that is, the position on the vertical axis in the right diagram of FIG. The axis position, that is, the position on the right and left axis in the right diagram of FIG. 5 is determined and plotted as the count number at the intersection CP.

Similarly, the three-dimensional object detection unit 33 defines lines Lb, Lc... In the direction in which the three-dimensional object falls, counts the number of difference pixels DP, and determines the horizontal axis position based on the position of each intersection CP. Then, the vertical axis position is determined from the count number (number of difference pixels DP) and plotted. The three-dimensional object detection unit 33 generates the differential waveform DW _t as shown in the right diagram of FIG.

Here, the difference pixel PD on the data of the difference image PD _t is a pixel that has changed in the images at different times, in other words, a location where a three-dimensional object exists. For this reason, the difference waveform DW _t is generated by counting the number of pixels along the direction in which the three-dimensional object collapses and performing frequency distribution at the location where the three-dimensional object exists. In particular, since the number of pixels is counted along the direction in which the three-dimensional object falls, the differential waveform DW _t is generated from the information in the height direction for the three-dimensional object.

As shown in the left diagram of FIG. 5, the line La and the line Lb in the direction in which the three-dimensional object collapses have different distances overlapping the detection area A1. For this reason, if the detection area A1 is filled with the difference pixels DP, the number of difference pixels DP is larger on the line La than on the line Lb. For this reason, when the three-dimensional object detection unit 33 determines the vertical axis position from the count number of the difference pixels DP, the three-dimensional object detection unit 33 is normalized based on the distance at which the lines La and Lb in the direction in which the three-dimensional object falls and the detection area A1 overlap. Turn into. As a specific example, in the left diagram of FIG. 5, there are six difference pixels DP on the line La, and there are five difference pixels DP on the line Lb. For this reason, in determining the vertical axis position from the count number in FIG. 5, the three-dimensional object detection unit 33 normalizes the count number by dividing it by the overlap distance. Thus, as shown in the difference waveform DW _t, the line La on the direction the three-dimensional object collapses, the value of the differential waveform DW _t corresponding to Lb is substantially the same.

After the generation of the difference waveform DW _t , the three-dimensional object detection unit 33 detects the adjacent vehicle existing in the adjacent lane based on the generated difference waveform DW _t . Here, FIG. 6 is a diagram for explaining a detection method of the adjacent vehicle by the three-dimensional object detection unit 33, and shows an example of the differential waveform DW _t and the threshold value α for detecting the adjacent vehicle. As illustrated in FIG. 6, the three-dimensional object detection unit 33 determines whether or not the peak of the generated differential waveform DW _t is greater than or _equal to a threshold value α corresponding to the peak position of the differential waveform DW _t . Then, when the peak of the differential waveform DW _t is less than the predetermined threshold value α, the three-dimensional object detection unit 33 determines that no adjacent vehicle exists in the detection areas A1 and A2, while the peak of the differential waveform DW _t Is equal to or greater than a predetermined threshold value α, it is determined that there is an adjacent vehicle in the detection areas A1 and A2, thereby detecting the adjacent vehicle existing in the adjacent lane.

As described above, the differential waveform DW _t is an aspect of pixel distribution information indicating a predetermined luminance difference, and the “pixel distribution information” in the present embodiment is obtained when the captured image is converted into a bird's-eye view image. It can be positioned as information indicating the distribution state of “pixels having a luminance difference equal to or greater than a predetermined threshold” detected along the direction in which the three-dimensional object falls. In other words, the three-dimensional object detection unit 33 distributes pixels whose luminance difference is equal to or greater than the predetermined threshold th in the direction in which the three-dimensional object falls when the viewpoint is converted into the bird's-eye view image on the bird's-eye view image obtained by the viewpoint conversion unit 31. When the information is detected as the difference waveform DW _{t and} the degree of pixel distribution in the direction in which the three-dimensional object collapses (the count number of the difference pixels DP in the difference waveform DW _t ) is equal to or greater than the threshold value α, the difference waveform DW _t Based on this, a three-dimensional object is detected.

Subsequently, a method for setting the threshold value α for detecting the adjacent vehicle will be described.

The threshold value α is set by the detection reference setting unit 35 shown in FIG. In the present embodiment, the detection criterion setting unit 35 is set according to the rear distance from the camera 10 based on the certainty factor detected by the certainty factor determination unit 34 shown in FIG. One of the threshold value α ₁ and the second threshold value α ₂ set according to the positional relationship between the camera 10 and the light source is set as a threshold value α for detecting the adjacent vehicle based on the differential waveform DW _t. To do.

When the light source is detected on the rear side of the host vehicle, the certainty factor determination unit 34 travels in the adjacent adjacent lane on the rear side of the host vehicle (the lane adjacent to the second lane of the host vehicle). The certainty of being a headlight of a vehicle (hereinafter referred to as an adjacent vehicle) is determined as a certainty factor. Below, the determination method of the certainty factor by the certainty factor determination part 34 is demonstrated. In this embodiment, as will be described later, detection of the light source is performed by the detection reference setting unit 35.

In the present embodiment, the certainty factor determination unit 34 determines the certainty factor that the detected light source is the headlight of the adjacent vehicle based on the number of detected light sources, the position of the light source, and the temporal change in the state of the light source. judge.

For example, when the number of light sources detected in the detection areas A1 and A2 is large, the certainty factor determination unit 34 is, for example, a street lamp or an illuminated signboard, and the head of the adjacent vehicle The possibility that the light is low is determined, and the certainty that the detected light source is the headlight of the adjacent vehicle is determined to be low.

In addition, the certainty determination unit 34 detects based on the position of the light source, that is, the height of the light source (the position of the light source in the vertical direction), the position of the light source in the vehicle width direction, and the position of the light source in the traveling direction of the host vehicle. The certainty factor that the light source that has been used is the headlight of the adjacent vehicle is determined. For example, when the height of the light source is high, the certainty factor determination unit 34 determines that the detected light source is highly likely to be caused by a streetlight, and determines the certainty factor that the detected light source is a headlight of an adjacent vehicle. Judge low. In addition, the certainty factor determination unit 34 uses the headlight of the adjacent vehicle that is located in the adjacent lane farther from the own vehicle than the adjacent lane as the position of the light source in the vehicle width direction is farther from the own vehicle. It is determined that there is a high possibility, and the certainty that the detected light source is the headlight of the adjacent vehicle is determined to be high. Further, the certainty determination unit 34 determines that the detected light source is less likely to be a headlight of an adjacent vehicle as the position of the light source in the traveling direction of the host vehicle is farther from the host vehicle. The certainty that is the headlight of the adjacent vehicle is determined low.

Furthermore, the certainty factor determination unit 34 determines the certainty factor that the detected light source is the headlight of the adjacent vehicle based on the temporal change in the state of the light source. For example, when the temporal change in the size of the detected light source is small, the certainty factor determination unit 34 uses a headlight of an adjacent vehicle that travels in an adjacent lane farther from the vehicle than the adjacent lane. It is determined that there is a high possibility, and the certainty factor that the detected light source is the headlight of the adjacent vehicle is determined to be high. Further, for example, the certainty factor determination unit 34 detects the detected light source from the own vehicle rather than the adjacent lane even when the temporal change in the position of the detected light source is small or the temporal change in the luminance of the detected light source is small. It is determined that there is a high possibility that it is a headlight of an adjacent vehicle that travels in a distant adjacent lane, and the certainty that the detected light source is a headlight of the adjacent vehicle is determined to be high.

Then, the certainty factor determination unit 34 comprehensively determines the certainty factor that the detected light source is the headlight of the adjacent vehicle based on the number of light sources, the position of the light sources, and the temporal change of the light source state. In this way, by comprehensively determining the number of light sources, the positions of the light sources, and the temporal changes in the state of the light sources, it is possible to appropriately determine whether or not the detected light sources are caused by the headlights of adjacent vehicles. Can be determined. The certainty factor determination unit 34 may be configured to take into account the size of the detected light source when determining the certainty factor. For example, the certainty factor determination unit 34 can be configured to determine the certainty factor low when the size of the detected light source greatly deviates from the size of a general headlight of an adjacent vehicle adjacent to the vehicle.

Then, the detection reference setting unit 35 determines whether or not the certainty level determined by the certainty level determination unit 34 is equal to or greater than a predetermined value. The first threshold value α ₁ set according to the rear distance is set as the threshold value α for detecting the adjacent vehicle based on the differential waveform DW _t . On the other hand, when the certainty factor is greater than or equal to a predetermined value, A second threshold value α ₂ set according to the positional relationship between the camera 10 and the light source, which will be described later, is set as a threshold value α for detecting an adjacent vehicle based on the differential waveform DW _t . Here, first, a method of setting the first threshold value α ₁ according to the rear distance from the camera 10 will be described.

The detection reference setting unit 35 detects positions in the detection areas A1, A2 corresponding to the lines La, Lb, Lc... In the direction in which the three-dimensional object shown in the left diagram of FIG. 5 falls (hereinafter, detection positions in the detection areas A1, A2). For each of the detection positions in the detection areas A1, A2 (for example, the average luminance of the difference pixels DP on the lines La, Lb, Lc... In the direction in which the three-dimensional object shown in the left diagram of FIG. And according to the back distance from the camera 10, 1st threshold value (alpha) ₁ is set. Here, FIG. 7 is a graph showing an example of the relationship between ₁ brightness and the first threshold value α at the detection position of the detection area A1, A2, 8, from the camera 10 rear distance and the first It is a figure for demonstrating the relationship with threshold value (alpha) ₁ .

Specifically, the detection reference setting unit 35, as shown in FIG. 7, the higher the brightness at the detection position of the detection area A1, A2, the first threshold value alpha ₁ corresponding to the detected position to a higher value Set. Thereby, for example, even when the headlight of the adjacent vehicle with high brightness (the vehicle existing in the adjacent adjacent lane adjacent to the second lane of the own lane) is irradiated on the adjacent lane, the light of the headlight of the adjacent vehicle Since the peak of the difference waveform DW _t based on the value becomes less than the first threshold value α ₁ and the influence of the light of the headlight of the adjacent vehicle adjacent to the vehicle can be eliminated, Can be effectively prevented from being erroneously detected.

Further, the detection reference setting unit 35, based on the first threshold value alpha ₁ of the gain that is set in accordance with the rearward distance from the camera 10 is set in accordance with the luminance at the detection position of the detection area A1, A2 to the change the first threshold value alpha ₁ was. For example, the detection reference setting unit 35 divides the detection areas A1 and A2 into three areas according to the rear distance from the camera 10, as shown in the left diagram of FIG. In the example shown in the left diagram of FIG. 8, the detection reference setting unit 35 includes the detection areas A1 and A2 in a region R1 where the rear distance from the camera 10 is less than the distance D1, and the rear distance from the camera 10 is the distance D1 or more. The region R2 is divided into three regions: a region R2 less than the distance D2 and a region R3 whose rear distance from the camera 10 is equal to or greater than the distance D2 and less than the distance D3. Note that FIG. 8 is illustrated and described only the detection area A1, the first gain threshold alpha ₁ Similarly, the detection area A2 is set.

As shown in FIG. 8 right figure, the first threshold value alpha ₁ of the gain is set in accordance with the rearward distance from the camera 10. For example, among the detection area A1, the rear distance from the camera 10 is not less distance D1 above, and, in the region is less than the distance D2 R2, the first threshold value alpha ₁ of the gain is set to a constant value. In contrast, among the detection area A1, in the region R1 rear distance is less than the distance D1 from the camera 10, has a higher gain than the first threshold value alpha ₁ that is set in the region R2, also, more rearward distance from the camera 10 is small (closer to the host vehicle), the first gain threshold alpha ₁ is higher. Furthermore, among the detection area A1, the rear distance from the camera 10 even distance D2 or more regions R3, has a higher gain than the first threshold value alpha ₁ that is set in the region R2, also from the camera 10 the larger the rearward distance (farther from the vehicle), the first gain threshold alpha ₁ is higher.

Here, FIG. 9A shows a scene in which the headlight of the adjacent vehicle V3 is detected in the region R2 in the detection region A2. As shown in FIG. 9A, in the region R2 of the detection region A2, the light of the headlight of the adjacent vehicle is irradiated linearly from the adjacent vehicle to the camera 10, and the light amount of the headlight incident on the camera 10 is Therefore, the luminance in the region R2 is higher than the other regions R1 and R2 in the detection region A2. Therefore, as shown in FIG. 8 right figure, in the region R2 in the detection areas A1, A2, the first gain threshold alpha _1, other regions R1, R3 as compared to a low value in the detection areas A1, A2 Even in this case, the detection reference setting unit 35 can set the threshold value α high in accordance with the high luminance in the region R2, as shown in FIG. 7, and thereby the head of the adjacent vehicle V3 in the region R2 can be set. It is possible to effectively prevent light from being erroneously detected as an adjacent vehicle. Note that FIG. 9A is illustrated with only the detection region A2, but the same applies to the detection region A1 (the same applies to FIGS. 9B and 9C described later).

FIG. 9B shows a scene in which the headlight of the adjacent vehicle V3 is detected in the region R1 within the detection region A2. As shown in FIG. 9B, when the headlight of the adjacent vehicle V3 is detected in the region R1 close to the host vehicle, the headlight light emitted from the adjacent vehicle V3 is in the direction in which the host vehicle V1 exists. Only a part of the irradiated light enters the camera 10. For this reason, in the region R1 close to the host vehicle in the detection regions A1 and A2, the amount of headlight incident on the camera 10 is usually smaller and the luminance of the region R1 is lower than the region R2 in the detection regions A1 and A2. Become. Therefore, as shown in FIG. 7, only changed according the first threshold alpha ₁ to the luminance, the light of headlights next adjacent vehicle V3 irradiated to the region R1, it is erroneously detected as the adjacent vehicle There is a case. In particular, in the region R1, as the rear distance from the camera 10 is smaller (closer to the host vehicle), the light of the headlight emitted from the adjacent vehicle V3 becomes weaker. In contrast, in the present embodiment, as shown in FIG. 8 right figure, in the region R1, as compared to the area R2, together with the first gain threshold alpha ₁ is set higher, from the camera 10 more rear distance is smaller (closer to the host vehicle), the first gain threshold alpha ₁ is set high. Therefore, in the present embodiment, the detection reference setting unit 35 can change the threshold value α to a high value even when the brightness of the headlight of the adjacent vehicle V3 detected in the region R1 is low. Even when the brightness of the headlight of the adjacent vehicle V3 detected at R1 is low, it is possible to effectively prevent the light of the headlight of the adjacent vehicle V3 from being erroneously detected as the adjacent vehicle V2.

FIG. 9C shows a scene in which the headlight of the adjacent vehicle V3 is detected in the region R3 in the detection region A2. As shown in FIG. 9C, a region R3 far from the host vehicle has a plurality of vehicles positioned behind the detection regions A1 and A2 (for example, the adjacent vehicle V3 shown in FIG. 9C and the adjacent vehicle V3. Furthermore, light of various luminances is emitted from a vehicle or the like existing behind. That is, the light of the headlight of the adjacent vehicle V3 detected in the region R3 may be lower in brightness than the region R2, and in such a case, as shown in FIG. 7, the first threshold value α ₁ May be erroneously detected as an adjacent vehicle if the image of the headlight of the adjacent vehicle V3 irradiated to the region R3 is simply changed according to the luminance. In particular, in the region R3, as the rear distance from the camera 10 increases (the farther from the host vehicle), the luminance of the light of the headlight to be irradiated tends to vary, and there is a possibility that light with lower luminance is irradiated. is there. In contrast, in the present embodiment, as shown in FIG. 8 right figure, in the area R3, similarly to the region R1, together with the first gain threshold alpha ₁ than a region R2 is set to be higher , more rearward distance from the camera 10 is larger (farther from the vehicle), the first gain threshold alpha ₁ is set high. Therefore, in the present embodiment, the detection reference setting unit 35 can be changed even if the brightness of the headlight of the next adjacent vehicle V3 detected in region R3 is low, the first threshold value alpha ₁ to a higher value, Thereby, even when the brightness of the headlight of the adjacent vehicle V3 detected in the region R3 is low, it is effectively prevented that the light of the headlight of the adjacent vehicle V3 is erroneously detected as the adjacent vehicle V2. Can do.

Thus, the detection reference setting unit 35, as shown in FIG. 7, sets a first threshold value alpha ₁ according to the luminance at the detection position of the detection area A1, A2, the first threshold set alpha The first threshold corresponding to each detection position in the detection areas A1 and A2 as shown in FIG. 6 by changing the gain of ₁ according to the rear distance from the camera 10 as shown in FIG. Set α ₁ . For example, referring to FIG. 7, the detection reference setting unit 35 sets the first threshold value α ₁ according to the luminance of the difference pixel DP on the line La shown in the left diagram of FIG. of the first threshold value alpha ₁ of the gain shown in, the first threshold value alpha ₁ of the gain at the intersections CP line La, by changing the threshold value alpha ₁ threshold first set, corresponding to line La calculating a first threshold value alpha ₁ in the position. Similarly, the detection reference setting unit 35, the line Lb, by calculating the first threshold value alpha ₁ at the position corresponding to Lc ..., as shown in FIG. 6, the respective detection positions in the detection areas A1, A2 setting the first threshold value alpha ₁ corresponding.

The detection reference setting unit 35 sets a threshold value α so as to prevent the headlight light image of the adjacent vehicle V3 irradiated on the adjacent lane from being erroneously detected as an adjacent vehicle. It is. Therefore, in this embodiment, the first threshold value alpha ₁ for changing the processing depending on the rearward distance from the camera 10, the conditions next adjacent vehicle V3 is headlights (e.g., at night) may be performed only . For example, when the luminance of the image captured by the camera 10 is equal to or lower than a predetermined value, the detection reference setting unit 35 determines that the adjacent adjacent vehicle V3 is in a condition for lighting the headlight (for example, at night). Can do.

Next, the second threshold value alpha ₂ setting method in accordance with the positional relationship between the camera 10 and the light source will be described. In the present embodiment, as shown in FIG. 10, the detection reference setting unit 35 is a first threshold map showing the relationship between the luminance at the detection positions in the detection areas A1 and A2 and the predetermined threshold α ₂ ′. As shown in FIG. 11, a second threshold map showing the relationship between the luminance at the detection positions in the detection areas A1 and A2 and the predetermined threshold α ₂ ″ is provided. The first threshold shown in FIG. In the threshold value map, when the adjacent vehicle is detected, the threshold value α ₂ ′ is relatively low when the brightness is relatively low so that the tire wheel which is a characteristic part of the adjacent vehicle is easily detected. On the other hand, the second threshold map shown in Fig. 11 is, for example, the head of an adjacent vehicle (a vehicle existing in the adjacent adjacent lane adjacent to the two lanes in the own lane) in the detection areas A1 and A2. Detect even if light is detected In headlights around that, so as not to cause erroneous detection of the adjacent vehicle, is set to be a relatively high value threshold alpha _{2 "even} when the luminance is relatively low. Specifically, compared to the threshold the threshold alpha _{2 'as} defined in the first threshold value map shown in FIG. 10, a second threshold value alpha _{2 ",} which is defined in the threshold map of shown in FIG. 11, luminance even if low, is set so as not to be below a predetermined value S _t.

Then, the detection reference setting unit 35 detects the threshold value α ₂ ′ obtained from the first threshold map shown in FIG. 10 according to the luminance at the detection position in the detection regions A1 and A2, and the detection in the detection regions A1 and A2. Using the threshold value α ₂ ″ obtained from the second threshold map shown in FIG. 11 in accordance with the luminance at the position, the adjacent vehicle is detected at the detection positions in the detection areas A1 and A2 according to the following formula 1. calculating a second threshold value alpha ₂ for.
[Equation 1]
Second threshold value α ₂ = {(1-wt) · threshold value α ₂ ′ obtained from the first threshold map) + (wt · threshold value α ₂ ″ obtained from the second threshold map)
In the above equation 1, wt is a weight of the threshold α ₂ ″ obtained from the second threshold map (a map that emphasizes prevention of erroneous detection around the headlight), and this weight wt is shown in FIG. It is determined based on the control map shown in FIG.

Here, FIG. 12 is an example of a control map that defines the weight wt of the threshold value α ₂ ″ obtained from the second threshold map. In the control map shown in FIG. 12, the vertical axis represents the second threshold value. The weight α of the threshold value α ₂ ″ obtained from the map is shown, and the position of the detected position in the traveling direction of the host vehicle is shown on the horizontal axis. In the control map shown in FIG. 12, the reference position P _O to be described later, weight wt is "1" up. Therefore, when calculating the second threshold value alpha ₂ according to the above equation 1, the second threshold value alpha ₂ at the reference position P _O is obtained from the second threshold value map that emphasizes prevention of erroneous detection at around headlights equal to the threshold value alpha _{2 "to} be. Further, in the control map shown in FIG. 12, the weight wt also in the vicinity of the reference position P _O is a value close to" 1 ". Therefore, when calculating the second threshold value alpha ₂ according to the above equation 1, the second threshold value alpha ₂ in the vicinity of the reference position P _O is a value close to the threshold alpha _{2 "obtained} from the second threshold value map.

In the control map shown in FIG. 12, the weight wt is “0” at a position slightly behind the reference position P _O behind the reference position P _O compared to the front of the reference position P _O. Therefore, when calculating the second threshold value alpha ₂ according to the above equation 1, the reference position P second threshold alpha ₂ in the rear of _O is obtained from the first threshold value map easily detect the tire wheel of the adjacent vehicle It becomes equal to the threshold value α ₂ ′. On the contrary, in front of the reference position _PO , the weight wt of the threshold value α ₂ ″ obtained from the second threshold map changes with a high value. Therefore, the second threshold value α ₂ is calculated according to the above equation 1. when, in front of the reference position P _O, as compared to the rear reference position P _O, so that the threshold value alpha _{2 "with} an emphasis on preventing the false detection of around headlights is calculated.

Furthermore, in the present embodiment, the detection reference setting unit 35 uses the control map shown in FIG. 12 before calculating the second threshold value α ₂ for detecting the adjacent vehicle. The adjustment is made according to the positional relationship between the camera 10 and the light source. The control map adjustment method shown in FIG. 12 will be described below with reference to FIG. Here, FIG. 13 is a diagram for explaining a method of adjusting the control map shown in FIG.

That is, as shown in the right diagram of FIG. 13A, the detection reference design unit 35 first detects a light source such as a headlight of the adjacent vehicle V3 from the captured image captured by the camera 10. Then, the detection reference setting unit 35 detects the position of the center of gravity of the detected light source, and sets a straight line Lc passing through the position of the center of gravity of the light source and the center position of the camera 10. Further, the detection reference setting unit 35 calculates an intersection O between the straight line Lc and a side (side along the traveling direction) L2 'on the adjacent adjacent lane side of the detection area A2. A method of detecting the light source by the detection reference setting unit 35 will be described later.

Then, the detection reference setting unit 35, as shown in FIG. 13 (A) left, and intersection point O, so that the reference position P _O of the control map shown in FIG. 12 coincide, the control map shown in FIG. 12 Make adjustments. Here, FIG. 13B illustrates a scene in which the adjacent vehicle V3 is close to the host vehicle V1 from the scene illustrated in FIG. In the scene shown in FIG. 13B, since the adjacent vehicle V3 is close to the host vehicle V1, the intersection O between the straight line Lc and the line L2 ′ of the detection area A2 is the intersection O shown in FIG. Rather than forward (Y-axis negative direction). Therefore, as shown in FIG. 13 (B) left, the detection reference setting unit 35, the position of the intersection point O shown in FIG. 13 (B) right figure, and the reference position P _O of the control map shown in FIG. 12 coincide Thus, by adjusting the control map shown in FIG. 12, the control map shown in FIG. 12 is compared with the control map shown in the left figure of FIG. are generally shifted forward (Y-axis negative direction), thereby, the position of the light source, and the reference position P _O of the control map shown in FIG. 12 so that the corresponding.

Thus, the reference position P _O of the control map shown in position and Figure 12 of the light source is made to correspond, the front region than the light source Rf (i.e., the traveling direction of the vehicle Y direction, the vehicle width direction X In the XY plane, the position of the camera 10 in the Y direction is 0, the rear of the camera 10 is Y> 0, and the straight line Lc is represented by Y = kX. The weight wt of the threshold value ₂ ″ obtained from the threshold value map of 2 can be increased. Therefore, when the second threshold value α ₂ is calculated according to the above equation 1, the area around the headlight in the area ahead of the light source Thus, the second threshold value α ₂ that emphasizes prevention of erroneous detection at the point is obtained.

On the other hand, the reference position P _O of the control map shown in position and Figure 12 of the light source is made to correspond, the rear area than the light source Rr (i.e., the traveling direction of the vehicle Y direction, the vehicle width direction and the X direction In the XY plane, when the position of the camera 10 in the Y direction is 0, the rear side of the camera 10 is Y> 0, and the straight line Lc is represented by Y = kX, the second region is Y ≧ kX). The weight wt of the threshold value α ₂ ″ obtained from the threshold map can be reduced. For this reason, when the second threshold value α ₂ is calculated according to the above equation 1, in the region behind the light source, the tire of the adjacent vehicle The second threshold value α ₂ that can be easily detected by the wheel is obtained.

The detection reference setting unit 35, the light of headlights next adjacent vehicle to be irradiated on the adjacent lane, in order to effectively prevent the erroneous detection as the adjacent vehicle, changing the second threshold value alpha ₂ To do. Therefore, in this embodiment, it can be set as the structure which performs the change of the 2nd threshold value (alpha) ₂ by the detection reference setting part 35 only at night when an adjacent vehicle lights a headlight. Note that the detection reference setting unit 35 may determine that it is nighttime, for example, when the luminance of an image captured by the camera 10 is equal to or less than a predetermined value.

As described above, the detection reference setting unit 35 has the first threshold value α ₁ set according to the rear distance from the camera 10 and the second threshold value α ₂ set according to the positional relationship between the camera 10 and the light source. And can be set. Then, the detection criterion setting unit 35 determines whether or not the certainty factor determined by the certainty factor determination unit 34 is equal to or greater than a predetermined value. If the certainty factor is less than the predetermined value, the first threshold value α ₁ is set as a threshold value α for detecting an adjacent vehicle based on the differential waveform DW _t . On the other hand, when the certainty factor is equal to or greater than a predetermined value, the second threshold value α _{2 is set} to the differential waveform DW _t . Based on this, a threshold value α for detecting the adjacent vehicle is set.

Subsequently, the description returns to the description of the three-dimensional object detection unit 33. After detecting the three-dimensional object present in the adjacent lane, the three-dimensional object detection unit 33, in contrast with the differential waveform DW _t in one time before the differential waveform DW _t-1 and the current time, of the three-dimensional object present on the adjacent vehicle Calculate the travel distance. That is, the three-dimensional object detection unit 33 calculates the movement distance from the time change of the difference waveforms DW _t and DW _t−1 .

More specifically, as shown in FIG. 14, the three-dimensional object detection unit 33 divides the differential waveform DW _t into a plurality of small areas DW _t1 to DW _tn (n is an arbitrary integer equal to or greater than 2). FIG. 14 is a diagram illustrating the small areas DW _t1 to DW _tn divided by the three-dimensional object detection unit 33. The small areas DW _t1 to DW _tn are divided so as to overlap each other, for example, as shown in FIG. For example, the small area DW _t1 and the small area DW _t2 overlap, and the small area DW _t2 and the small area DW _t3 overlap.

Next, the three-dimensional object detection unit 33 obtains an offset amount (amount of movement of the differential waveform in the horizontal axis direction (vertical direction in FIG. 14)) for each of the small regions DW _t1 to DW _tn . Here, the offset amount is determined from the difference between the differential waveform DW _t in the difference waveform DW _t-1 and the current time before one unit time (distance in the horizontal axis direction). At this time, three-dimensional object detection unit 33, for each small area _DW t1 ~ _{DW tn,} when moving the differential waveform _{DW t1} before one unit time in the horizontal axis direction, the differential waveform DW _t at the current time The position where the error is minimized (the position in the horizontal axis direction) is determined, and the amount of movement in the horizontal axis between the original position of the differential waveform DW _t−1 and the position where the error is minimized is obtained as an offset amount. Then, the three-dimensional object detection unit 33 counts the offset amount obtained for each of the small areas DW _t1 to DW _{tn and forms} a histogram.

FIG. 15 is a diagram illustrating an example of a histogram obtained by the three-dimensional object detection unit 33. As shown in FIG. 15, the offset amount, which is the amount of movement that minimizes the error between each of the small areas DW _t1 to DW _tn and the difference waveform DW _t−1 one time before, has some variation. For this reason, the three-dimensional object detection unit 33 forms a histogram of offset amounts including variations, and calculates a movement distance from the histogram. At this time, the three-dimensional object detection unit 33 calculates the moving distance of the adjacent vehicle from the maximum value of the histogram. That is, in the example illustrated in FIG. 15, the three-dimensional object detection unit 33 calculates the offset amount indicating the maximum value of the histogram as the movement distance τ ^* . As described above, in this embodiment, even if the offset amount varies, it is possible to calculate a more accurate movement distance from the maximum value. The moving distance τ ^* is a relative moving distance of the adjacent vehicle with respect to the host vehicle V1. For this reason, when calculating the absolute movement distance, the three-dimensional object detection unit 33 calculates the absolute movement distance based on the obtained movement distance τ ^* and the signal from the vehicle speed sensor 20.

Thus, in the present embodiment, a one-dimensional waveform is obtained by calculating the moving distance of the three-dimensional object from the offset amount of the differential waveform DW _t when the error of the differential waveform DW _t generated at different times is minimized. Thus, the movement distance is calculated from the offset amount of the information, and the calculation cost can be suppressed in calculating the movement distance. Further, by dividing the differential waveform DW _t generated at different times into a plurality of small areas DW _t1 to DW _tn , it is possible to obtain a plurality of waveforms representing respective portions of the three-dimensional object. Since the offset amount can be obtained for each of the positions, and the movement distance can be obtained from a plurality of offset amounts, the calculation accuracy of the movement distance can be improved. Further, in the present embodiment, by calculating the moving distance of the three-dimensional object from the time change of the differential waveform DW _t including the information in the height direction, compared with a case where attention is paid only to one point of movement, Since the detection location before the time change and the detection location after the time change are specified including information in the height direction, it is likely to be the same location in the three-dimensional object, and the movement distance will be calculated from the time change of the same location, The calculation accuracy of the movement distance can be improved.

Note that the three-dimensional object detection unit 33 weights each of the plurality of small areas DW _t1 to DW _{tn and forms} a histogram by counting the offset amount obtained for each of the small areas DW _t1 to DW _tn according to the weight. May be. FIG. 16 is a diagram illustrating weighting by the three-dimensional object detection unit 33.

As shown in FIG. 16, the small area DW _m (m is an integer of 1 to n−1) is flat. That is, in the small area DW _m , the difference between the maximum value and the minimum value of the number of pixels indicating a predetermined difference is small. Three-dimensional object detection unit 33 to reduce the weight for such small area DW _m. This is because the flat small area DW _m has no characteristics and is likely to have a large error in calculating the offset amount.

On the other hand, the small region DW _{m + k} (k is an integer equal to or less than nm) is rich in undulations. That is, in the small area DW _m , the difference between the maximum value and the minimum value of the number of pixels indicating a predetermined difference is large. Three-dimensional object detection unit 33 increases the weight for such small area DW _m. This is because the small region DW _{m + k} rich in undulations is characteristic and there is a high possibility that the offset amount can be accurately calculated. By weighting in this way, the calculation accuracy of the movement distance can be improved.

In the above embodiment, the differential waveform DW _t is divided into a plurality of small areas DW _t1 to DW _tn in order to improve the calculation accuracy of the movement distance. However, when the calculation accuracy of the movement distance is not so required, the small area DW _t1 is divided. It is not necessary to divide into ~ DW _tn . In this case, the three-dimensional object detection unit 33 calculates the moving distance from the offset amount of the differential waveform DW _t when the error between the differential waveform DW _t and the differential waveform DW _t−1 is minimized. That is, the method for obtaining the offset amount of the difference waveform DW _t in the difference waveform DW _t-1 and the current time before one unit time is not limited to the above disclosure.

In the present embodiment, the three-dimensional object detection unit 33 obtains the moving speed of the host vehicle V1 (camera 10), and obtains the offset amount for the stationary object from the obtained moving speed. After obtaining the offset amount of the stationary object, the three-dimensional object detection unit 33 ignores the offset amount corresponding to the stationary object among the maximum values of the histogram and calculates the moving distance of the adjacent vehicle.

FIG. 17 is a diagram illustrating another example of a histogram obtained by the three-dimensional object detection unit 33. When there is a stationary object in addition to the adjacent vehicle within the angle of view of the camera 10, two maximum values τ1 and τ2 appear in the obtained histogram. In this case, one of the two maximum values τ1, τ2 is the offset amount of the stationary object. For this reason, the three-dimensional object detection unit 33 calculates the offset amount for the stationary object from the moving speed, ignores the maximum value corresponding to the offset amount, and calculates the moving distance of the three-dimensional object by using the remaining maximum value. To do. Thereby, the situation where the calculation accuracy of the moving distance of a solid object falls by a stationary object can be prevented.

Even if the offset amount corresponding to a stationary object is ignored, if there are a plurality of maximum values, it is assumed that there are a plurality of adjacent vehicles within the angle of view of the camera 10. However, it is extremely rare that a plurality of adjacent vehicles exist in the detection areas A1 and A2. For this reason, the three-dimensional object detection unit 33 stops calculating the movement distance. Thereby, in the present embodiment, it is possible to prevent a situation in which an erroneous movement distance having a plurality of maximum values is calculated.

Next, the adjacent vehicle detection process according to the present embodiment will be described. FIG. 18 is a flowchart showing the adjacent vehicle detection process of the present embodiment. As shown in FIG. 18, first, the computer 30 acquires data of the captured image P from the camera 10 (step S101), and the viewpoint conversion unit 31 performs bird's-eye view based on the acquired data of the captured image P. Data of the image PB _t is generated (step S102).

Next, the alignment unit 33 aligns the data of the bird's-eye view image PB _{t and} the data of the bird's-eye view image PB _{t−1 one} hour before, and generates data of the difference image PD _t (step S103). . Then, three-dimensional object detection unit 33, from the data of the difference image PD _t, pixel value by counting the number of difference pixel DP "1", to generate a difference waveform DW _t (step S104).

The detection reference setting unit 35 sets a threshold value α for detecting an adjacent vehicle based on the differential waveform DW _t (step S105). Here, FIG. 19 is a flowchart showing the threshold value α setting processing in step S105. As shown in FIG. 19, the detection reference setting unit 35 first detects a light source such as a headlight of an adjacent vehicle from a captured image captured by the camera 10 (step S201).

Here, the detection reference setting unit 35 eliminates the influence of noise and appropriately detects a headlight of another vehicle as a light source, so that the difference in brightness from the surroundings in the captured image is equal to or greater than a predetermined value s _d And an image area having a size equal to or larger than the predetermined value s ₁ is detected as an area corresponding to the light source.

Therefore, the detection reference setting unit 35 first performs edge processing of the captured image, and detects a region where the brightness difference from the surroundings is equal to or greater than the predetermined value _sd as a light source candidate. In the present embodiment, the detection reference setting unit 35 does not limit the predetermined value _sd to a constant value in order to appropriately detect the light source, for example, the rear distance from the camera 10 to the light source candidate, The predetermined value s _d can be changed based on the brightness of the region where the candidate exists. For example, the detection reference setting unit 35 has a threshold map in which a predetermined value s _d is set according to the luminance, and a threshold map in which the predetermined value s _d is set according to the rear distance from the camera 10 to the light source candidate. and compares these two thresholds map, that the predetermined value s _d of the higher of the prescribed value s _d obtained for each threshold map is selected as the predetermined value s _d for detecting the light source candidate it can.

Then, the detection reference setting unit 35 detects an image area having a size equal to or larger than the predetermined value s ₁ among the detected light source candidates as an area corresponding to the light source. The predetermined value s _l is not limited to a certain value, for example, detection reference setting unit 35, in accordance with the rearward distance from the camera 10 to the light source candidates, may be changed a predetermined value s _l. For example, when the length of the detection areas A1 and A2 in the traveling direction of the host vehicle is 10 m, the detection reference setting unit 35 moves the detection areas A1 and A2 from the position closest to the camera 10 in the detection areas A1 and A2. The vehicle is divided into three regions, ie, a region R1 from 0 to 1.5 m, a region R2 from 1.5 to 6 m, and a region R3 from 6 m to 10 m in the vehicle traveling direction. Then, the detection reference setting unit 35 includes, for example, an image area in which the height × width is 5 × 5 pixels or more in the area R1 close to the own vehicle and the area R3 far from the own vehicle in the detection areas A1 and A2. For example, in the central region R2 of the detection regions A1 and A2, an image region having a height × width of 7 × 7 pixels or more is detected as a region corresponding to the light source. it can.

If the light source cannot be detected in step S201, the detection reference setting unit 35 sets the threshold α to a predetermined threshold, for example, and ends the threshold α setting processing in step S105.

Next, the certainty factor determination unit 34 determines the certainty factor that the detected light source is the headlight of the adjacent vehicle adjacent to the rear side of the host vehicle (step S202). As described above, the certainty factor determination unit 34 is convinced that the detected light source is the headlight of the adjacent vehicle based on the number of light sources, the position of the light sources, the temporal change in the state of the light sources, and the size of the light sources. Determining the degree comprehensively. Then, the detection reference setting unit 35 determines whether or not the certainty factor is equal to or greater than a predetermined value based on the determination result of the certainty factor in step S202. In order to set the threshold value α according to the rear distance from 10, the process proceeds to step S204, and when the certainty factor is less than the predetermined value, the threshold value α is set based on the positional relationship between the camera 10 and the light source. Then, the process proceeds to step S208.

In step S204, the detection reference setting unit 35, as shown in FIG. 7, in accordance with the detection position brightness, the first acquisition threshold alpha ₁ at the detection position is performed. Then, the detection reference setting unit 35 uses the first threshold value α ₁ acquired in step S204 as the gain of the first threshold value α ₁ defined according to the rear distance from the camera 10, as shown in FIG. Is changed based on (step S205). Then, the detection reference setting unit 35 first calculates a threshold value alpha ₁ for every detection position in the detection area A1, A2, calculates a first threshold value alpha ₁ at all the detection positions in the detection area A1, the A2 when (step S206 = Yes), the calculated first threshold value alpha ₁ was, as shown in FIG. 6, it is set as the threshold value alpha for detecting adjacent vehicle based on the difference waveform DW _t.

On the other hand, if it is determined in step S203 that the certainty factor is less than the predetermined value, the process proceeds to step S208. In step S208, the detection reference setting unit 35 sets a straight line Lc that passes through the center position of the detected light source and the center position of the camera 10, as shown in the right diagram of FIG. 35 calculates an intersection O between the set straight line Lc and the side L2 ′ on the adjacent adjacent lane side of the detection area A2 (step S209). Then, the detection reference setting unit 35, as shown in FIG. 13 (B) left, so that the intersection point O calculated in step S209, and the reference position P _O of the control map shown in FIG. 12 coincides, 12 Is adjusted (step S210).

Then, the detection criterion setting unit 35 detects the first threshold map (see FIG. 7) in which the threshold α ₂ ′ is set so that the tire wheel of the adjacent vehicle is easily detected, and erroneous detection around the headlight. Emphasizing prevention, the _second threshold value α ₂ ″ in which the threshold value α ₂ ″ is set (see FIG. 8) and the control map adjusted in step S210 are used, and the second threshold value α ₂ according to the above equation 1. is calculated (step S211). then, the detection reference setting unit 35, a second threshold value alpha ₂ calculates for every detection position in the detection area A1, A2, at all the detection positions in the detection area A1, the A2 when calculating the second threshold value alpha ₂ (step S212 = Yes), the second threshold alpha ₂ calculated, as shown in FIG. 6, the threshold value for detecting the adjacent vehicle based on the difference waveform DW _t Set as α (Step S213).

Returning to FIG. 18, the three-dimensional object detection unit 33 determines whether or not the peak of the differential waveform DW _t generated in step S104 is _equal to or greater than the threshold value α generated in step S105 (step S106). When the peak of the difference waveform DW _t is not equal to or greater than the threshold value α, that is, when there is almost no difference, it is considered that there is no adjacent vehicle in the captured image P. For this reason, when it is determined that the peak of the difference waveform DW _t is not equal to or greater than the threshold value α (step S106 = No), the three-dimensional object detection unit 33 determines that there is no adjacent vehicle in the adjacent lane (step S115). The process shown in FIG. 18 ends.

On the other hand, when it is determined that the peak of the differential waveform DW _t is equal to or greater than the threshold value α (step S106 = Yes), the three-dimensional object detection unit 33 determines that a three-dimensional object exists in the adjacent lane, and proceeds to step S107. The three-dimensional object detection unit 33 divides the differential waveform DW _t into a plurality of small areas DW _t1 to DW _tn . Next, the three-dimensional object detection unit 33 performs weighting for each of the small areas DW _t1 to DW _tn (step S108), calculates an offset amount for each of the small areas DW _t1 to DW _tn (step S109), and adds the weights. A histogram is generated (step S110).

Then, the three-dimensional object detection unit 33 calculates a relative movement distance that is a movement distance of the adjacent vehicle with respect to the own vehicle based on the histogram (step S111). Furthermore, the three-dimensional object detection unit 33 calculates the absolute movement speed of the adjacent vehicle from the relative movement distance (step S112). At this time, the three-dimensional object detection unit 33 calculates the relative movement speed by differentiating the relative movement distance with respect to time, and adds the own vehicle speed detected by the vehicle speed sensor 20 to calculate the absolute movement speed of the adjacent vehicle.

Thereafter, the three-dimensional object detection unit 33 determines whether the absolute movement speed of the adjacent vehicle is 10 km / h or more and the relative movement speed of the adjacent vehicle with respect to the own vehicle is +60 km / h or less (step S113). When both are satisfied (step S113 = Yes), the three-dimensional object detection unit 33 determines that there is an adjacent vehicle in the adjacent lane (step S114). Then, the process shown in FIG. 18 ends. On the other hand, when either one is not satisfied (step S113 = No), the three-dimensional object detection unit 33 determines that there is no adjacent vehicle in the adjacent lane (step S115). Then, the process shown in FIG. 18 ends.

In the present embodiment, the rear sides of the host vehicle are set as detection areas A1 and A2, and emphasis is placed on whether or not there is a possibility of contact when the host vehicle changes lanes. For this reason, the process of step S113 is performed. That is, assuming that the system according to the present embodiment is operated on a highway, if the speed of the adjacent vehicle is less than 10 km / h, even if the adjacent vehicle exists, the host vehicle is required to change the lane. Because it is located far behind, there are few problems. Similarly, when the relative moving speed of the adjacent vehicle to the own vehicle exceeds +60 km / h (that is, when the adjacent vehicle is moving at a speed higher than 60 km / h than the speed of the own vehicle), when changing the lane Is less likely to be a problem because it is moving in front of the host vehicle. For this reason, it can be said that the adjacent vehicle which becomes a problem at the time of lane change is judged in step S113.

Also, in step S113, determining whether the absolute moving speed of the adjacent vehicle is 10 km / h or more and the relative moving speed of the adjacent vehicle with respect to the own vehicle is +60 km / h or less has the following effects. For example, depending on the mounting error of the camera 10, the absolute moving speed of the stationary object may be detected to be several km / h. Therefore, by determining whether the speed is 10 km / h or more, it is possible to reduce the possibility that the stationary object is determined to be an adjacent vehicle. Further, depending on the noise, the relative speed of the adjacent vehicle to the host vehicle may be detected as a speed exceeding +60 km / h. Therefore, the possibility of erroneous detection due to noise can be reduced by determining whether the relative speed is +60 km / h or less.

Furthermore, instead of the process in step S113, it may be determined that the absolute movement speed of the adjacent vehicle is not negative or not 0 km / h. Further, in this embodiment, since an emphasis is placed on whether or not there is a possibility of contact when the host vehicle changes lanes, a warning sound is sent to the driver of the host vehicle when an adjacent vehicle is detected in step S114. Or a display corresponding to a warning may be performed by a predetermined display device.

As described above, in the first embodiment, the detection areas A1 and A2 behind the host vehicle are captured at different times, the captured images are converted into bird's-eye view images, and based on the difference between the bird's-eye view images at different times. The difference image PD _t is generated. Then, the difference waveform DW _t is generated from the data of the difference image PD _t by counting the number of pixels indicating a predetermined difference and performing frequency distribution along the direction in which the three-dimensional object falls by the viewpoint conversion. Then, it is determined whether or not the peak of the generated differential waveform DW _t is greater than or _equal to a rearward distance from the camera 10 or a threshold value α set according to the positional relationship between the camera 10 and the light source, and the differential waveform DW. _When the peak of _t is equal to or greater than the threshold value α, it is possible to appropriately detect the adjacent vehicle existing in the adjacent lane by determining that the adjacent vehicle exists in the adjacent lane.

That is, in this embodiment, when detecting an adjacent vehicle based on the differential waveform DW _t , the light source behind the host vehicle is detected, and the number of detected light sources, the position of the light source, the size of the light source, and Based on the temporal change in the state of the light source, the certainty factor that the detected light source is the headlight of the adjacent vehicle is determined. When the certainty factor is equal to or greater than a predetermined value, the second threshold value α ₂ obtained based on the positional relationship between the camera 10 and the light source is set as the threshold value α. That is, when there is a high possibility that the detected light source is a headlight of an adjacent vehicle, as shown in FIG. 10, the threshold value α is set in front of the light source to prevent erroneous detection around the headlight. The threshold value α is set so that the tire wheel of the adjacent vehicle is easily detected behind the light source. Thereby, for example, even when the brightness of the periphery of the headlight (light source) is high due to contamination of the lens of the camera 10 and the tire wheel of the adjacent vehicle, which is a characteristic part existing behind the headlight (light source), is difficult to detect. In front of the headlight (light source) that does not have a tire wheel, the threshold α is set to be high so as to prevent erroneous detection around the headlight. It can be effectively prevented, and in the rear of the headlight (light source), the threshold α is set so that the tire wheel of the adjacent vehicle can be easily detected, so it exists behind the headlight (light source). It is possible to properly detect the tire wheel of the adjacent vehicle.

Further, in the three-dimensional object detection device 1 according to the present embodiment, the first threshold map (see FIG. 10) in which the threshold value α ₂ ′ is set so as to easily detect the tire wheel of the adjacent vehicle, and the three-dimensional object by the light source. And a second threshold value map (see FIG. 11) in which a threshold value α ₂ ″ is set so as to prevent false detection, and when the certainty factor is equal to or greater than a predetermined value, The threshold value α ₂ ′ set and the threshold value α ₂ ″ set in the second threshold map are weighted according to the position from the light source with reference to the control map shown in FIG. Thus, a threshold value α for detecting the adjacent vehicle is calculated.

In the present embodiment, as the reference position P _O position control map shown in FIG. 12 of the light source coincides, by adjusting the control map shown in FIG. 12, the reference position corresponding to the position of the light source In the region ahead of _PO , the weight α of the threshold α ₂ ″ obtained from the second threshold map (threshold map that emphasizes prevention of erroneous detection of a three-dimensional object by the light source) becomes large, and the reference corresponding to the position of the light source in the region behind from the position P _O, weight wt of the second threshold value map threshold α2 obtained from (threshold map that emphasizes prevent erroneous detection of the three-dimensional object by a light source) "it is reduced. Thus, when the brightness is the same, the threshold value α for detecting the adjacent vehicle is higher in the area ahead of the light source, and the threshold value α for detecting the adjacent vehicle is lower in the area behind the light source. Become. Therefore, in this embodiment, the brightness of the periphery of the headlight (light source) is high due to dirt on the lens of the camera 10, and the tire wheel of the adjacent vehicle, which is a characteristic part existing behind the headlight (light source), is difficult to detect. Even in this case, it is possible to appropriately detect the tire wheel of the adjacent lane that exists behind the headlight (light source).

Further, when the light of the headlight (light source) of the adjacent vehicle is irradiated on the adjacent lane, the light of the headlight (light source) is reflected on the adjacent lane, and this reflected light is ahead of the headlight (light source). It may be detected in the area. Even in such a case, according to the present embodiment, since the threshold value α for detecting the adjacent vehicle can be increased with respect to the brightness in the area in front of the headlight (light source), such reflection is performed. It is possible to effectively prevent a light image from being erroneously detected as an adjacent vehicle.

On the other hand, when the certainty factor is less than the predetermined value, the first threshold value α ₁ obtained based on the rear distance from the camera 10 is set as the threshold value α. That is, when the possibility that the detected light source is the headlight of the adjacent vehicle is low, as shown in FIG. 8, in the detection areas A1 and A2, the area R1 close to the own vehicle and the area far from the own vehicle In R3, the gain of the threshold value α is set higher than that in the central region R2. Thereby, in the regions R1 and R3, the threshold value α can be set high even when the luminance of the headlight light emitted from the headlight from the adjacent vehicle is low, and the regions R1 and R3 are irradiated with the adjacent light. It is possible to effectively prevent erroneous detection of the light image of the headlight of the adjacent vehicle V3 as the adjacent vehicle.

Furthermore, in the present embodiment, in the case confidence is less than the predetermined value, the first threshold value alpha ₁ obtained based on the rearward distance from the camera 10 is set as the threshold alpha, confidence is greater than a predetermined value in some cases, setting the second threshold value alpha ₂ obtained on the basis of the positional relationship between the camera 10 and the light source as the threshold alpha. Here, when the certainty factor is greater than or equal to a predetermined value, that is, when the possibility that the detected light source is the headlight of the adjacent vehicle is high, the second obtained based on the positional relationship between the camera 10 and the light source. threshold alpha ₂ to be to the threshold alpha, when the detected light source is likely to be headlights next adjacent vehicle, the first threshold value alpha ₁ obtained based on the rearward distance from the camera 10 Compared with the case where the threshold value α is set, it is possible to eliminate the influence of the headlights of the adjacent vehicles and to obtain a higher effect of appropriately detecting the adjacent vehicles.

Further, when the certainty factor is less than a predetermined value, that is, when the possibility that the detected light source is a headlight of the adjacent vehicle is low, the second obtained based on the positional relationship between the camera 10 and the light source. when the threshold value alpha ₂ is set as the threshold alpha, for example, too much is set higher threshold alpha in the front than light, erroneous detection can not be properly detected neighboring vehicle when an adjacent vehicle actually exists It might happen. Therefore, in this embodiment, when the certainty factor is less than a predetermined value, the certainty factor is less than the predetermined value by setting the _first threshold value α1 obtained based on the rear distance from the camera 10 as the threshold value α. If it is, the second threshold value alpha ₂ obtained on the basis of the positional relationship between the camera 10 and the light source as compared with the case of setting the threshold value alpha, the erroneous detection of the adjacent vehicle with light of headlights next adjacent vehicle It can prevent more effectively and can detect an adjacent vehicle more appropriately.

<< Second Embodiment >>
Next, the three-dimensional object detection device 1a according to the second embodiment will be described. As shown in FIG. 20, the three-dimensional object detection device 1 a according to the second embodiment includes a computer 30 a instead of the computer 30 of the first embodiment, except that it operates as described below. This is the same as in the first embodiment. Here, FIG. 20 is a block diagram illustrating details of the computer 30a according to the second embodiment.

As shown in FIG. 20, the three-dimensional object detection device 1a according to the second embodiment includes a camera 10 and a computer 30a. The computer 30a includes a viewpoint conversion unit 31, a luminance difference calculation unit 36, and an edge line detection unit. 37, a three-dimensional object detection unit 33a, a certainty factor determination unit 34, and a detection reference setting unit 35a. Below, each structure of the solid-object detection apparatus 1a which concerns on 2nd Embodiment is demonstrated.

FIGS. 21A and 21B are diagrams illustrating an imaging range and the like of the camera 10 in FIG. 20. FIG. 21A is a plan view, and FIG. 21B is a perspective view in real space rearward from the host vehicle V1. Show. As shown in FIG. 21A, the camera 10 has a predetermined angle of view a, and images the rear side from the host vehicle V1 included in the predetermined angle of view a. Similarly to the case shown in FIG. 2, the angle of view a of the camera 10 is set so that the imaging range of the camera 10 includes the adjacent lane in addition to the lane in which the host vehicle V1 travels.

The detection areas A1 and A2 in this example are trapezoidal in a plan view (when viewed from a bird's eye), and the positions, sizes, and shapes of the detection areas A1 and A2 are determined based on the distances d ₁ to d _4. Is done. The detection areas A1 and A2 in the example shown in the figure are not limited to a trapezoidal shape, and may be other shapes such as a rectangle when viewed from a bird's eye view as shown in FIG.

Here, the distance d1 is a distance from the host vehicle V1 to the ground lines L1 and L2. The ground lines L1 and L2 mean lines on which a three-dimensional object existing in the lane adjacent to the lane in which the host vehicle V1 travels contacts the ground. In the present embodiment, the object is to detect adjacent vehicles V2 and the like (including two-wheeled vehicles) traveling in the left and right lanes adjacent to the lane of the host vehicle V1 on the rear side of the host vehicle V1. For this reason, a distance d1 which is a position to be the ground lines L1, L2 of the adjacent vehicle V2 is determined from a distance d11 from the own vehicle V1 to the white line W and a distance d12 from the white line W to a position where the adjacent vehicle V2 is predicted to travel. It can be determined substantially fixedly.

Further, the distance d1 is not limited to being fixedly determined, and may be variable. In this case, the computer 30a recognizes the position of the white line W with respect to the host vehicle V1 by a technique such as white line recognition, and determines the distance d11 based on the recognized position of the white line W. Thereby, the distance d1 is variably set using the determined distance d11. In the following embodiment, since the position where the adjacent vehicle V2 travels (distance d12 from the white line W) and the position where the host vehicle V1 travels (distance d11 from the white line W) are roughly determined, the distance d1 is It shall be fixedly determined.

The distance d2 is a distance extending in the vehicle traveling direction from the rear end portion of the host vehicle V1. The distance d2 is determined so that the detection areas A1 and A2 are at least within the angle of view a of the camera 10. In particular, in the present embodiment, the distance d2 is set so as to be in contact with the range divided into the angle of view a. The distance d3 is a distance indicating the length of the detection areas A1, A2 in the vehicle traveling direction. This distance d3 is determined based on the size of the three-dimensional object to be detected. In the present embodiment, since the detection target is the adjacent vehicle V2 or the like, the distance d3 is set to a length including the adjacent vehicle V2.

As shown in FIG. 21B, the distance d4 is a distance indicating a height set so as to include a tire such as the adjacent vehicle V2 in the real space. The distance d4 is a length shown in FIG. 21A in the bird's-eye view image. Note that the distance d4 may be a length that does not include a lane that is further adjacent to the left and right lanes in the bird's-eye view image (that is, the adjacent lane that is adjacent to two lanes). If the lane adjacent to the two lanes is included from the lane of the own vehicle V1, there is an adjacent vehicle V2 in the adjacent lane on the left and right of the own lane that is the lane in which the own vehicle V1 is traveling. This is because it becomes impossible to distinguish whether there is an adjacent vehicle on the lane.

As described above, the distances d1 to d4 are determined, and thereby the positions, sizes, and shapes of the detection areas A1 and A2 are determined. More specifically, the position of the upper side b1 of the detection areas A1 and A2 forming a trapezoid is determined by the distance d1. The starting point position C1 of the upper side b1 is determined by the distance d2. The end point position C2 of the upper side b1 is determined by the distance d3. The side b2 of the detection areas A1 and A2 having a trapezoidal shape is determined by a straight line L3 extending from the camera 10 toward the starting point position C1. Similarly, a side b3 of trapezoidal detection areas A1 and A2 is determined by a straight line L4 extending from the camera 10 toward the end position C2. The position of the lower side b4 of the detection areas A1 and A2 having a trapezoidal shape is determined by the distance d4. Thus, the areas surrounded by the sides b1 to b4 are set as the detection areas A1 and A2. As shown in FIG. 21B, the detection areas A1 and A2 are true squares (rectangles) in real space on the rear side from the host vehicle V1.

Referring back to FIG. 20, the viewpoint conversion unit 31 inputs captured image data of a predetermined area obtained by imaging with the camera 10. The viewpoint conversion unit 31 performs viewpoint conversion processing on the input captured image data to the bird's-eye image data in a bird's-eye view state. The bird's-eye view is a state seen from the viewpoint of a virtual camera looking down from above, for example, vertically downward (or slightly obliquely downward). This viewpoint conversion process can be realized by a technique described in, for example, Japanese Patent Application Laid-Open No. 2008-219063.

The luminance difference calculation unit 36 calculates a luminance difference with respect to the bird's-eye view image data subjected to viewpoint conversion by the viewpoint conversion unit 31 in order to detect the edge of the three-dimensional object included in the bird's-eye view image. For each of a plurality of positions along a vertical imaginary line extending in the vertical direction in the real space, the brightness difference calculation unit 36 calculates a brightness difference between two pixels in the vicinity of each position. The luminance difference calculation unit 36 can calculate the luminance difference by either a method of setting only one vertical virtual line extending in the vertical direction in the real space or a method of setting two vertical virtual lines.

Here, a specific method for setting two vertical virtual lines will be described. The brightness difference calculating unit 36 applies a first vertical imaginary line corresponding to a line segment extending in the vertical direction in the real space and a vertical direction in the real space different from the first vertical imaginary line with respect to the bird's eye view image that has undergone viewpoint conversion. A second vertical imaginary line corresponding to the extending line segment is set. The luminance difference calculation unit 36 continuously obtains the luminance difference between the point on the first vertical imaginary line and the point on the second vertical imaginary line along the first vertical imaginary line and the second vertical imaginary line. Hereinafter, the operation of the luminance difference calculation unit 36 will be described in detail.

As shown in FIG. 22A, the luminance difference calculation unit 36 corresponds to a line segment extending in the vertical direction in real space and passes through the detection area A1 (hereinafter, attention line La). Set). In addition, unlike the attention line La, the luminance difference calculation unit 36 corresponds to a line segment extending in the vertical direction in the real space and also passes through the detection area A1 and is a second vertical imaginary line Lr (hereinafter referred to as a reference line Lr). Set. Here, the reference line Lr is set at a position separated from the attention line La by a predetermined distance in the real space. Note that the line corresponding to the line segment extending in the vertical direction in the real space is a line that spreads radially from the position Ps of the camera 10 in the bird's-eye view image. This radially extending line is a line along the direction in which the three-dimensional object falls when converted to bird's-eye view.

The luminance difference calculation unit 36 sets a point of interest Pa (a point on the first vertical imaginary line) on the line of interest La. In addition, the luminance difference calculation unit 36 sets a reference point Pr (a point on the second vertical plate) on the reference line Lr. The attention line La, the attention point Pa, the reference line Lr, and the reference point Pr have the relationship shown in FIG. 22B in the real space. As apparent from FIG. 22B, the attention line La and the reference line Lr are lines extending in the vertical direction in the real space, and the attention point Pa and the reference point Pr are substantially the same height in the real space. This is the point that is set. Note that the attention point Pa and the reference point Pr do not necessarily have the same height, and an error that allows the attention point Pa and the reference point Pr to be regarded as the same height is allowed.

The luminance difference calculation unit 36 calculates a luminance difference between the attention point Pa and the reference point Pr. If the luminance difference between the attention point Pa and the reference point Pr is large, it is considered that an edge exists between the attention point Pa and the reference point Pr. In particular, in the second embodiment, in order to detect a three-dimensional object existing in the detection areas A1 and A2, a vertical virtual line is set as a line segment extending in the vertical direction in the real space with respect to the bird's-eye view image, In the case where the luminance difference between the attention line La and the reference line Lr is high, there is a high possibility that there is an edge of the three-dimensional object at the set position of the attention line La. For this reason, the edge line detection unit 37 shown in FIG. 20 detects an edge line based on the luminance difference between the attention point Pa and the reference point Pr.

This point will be explained in more detail. FIG. 23 is a diagram showing a detailed operation of the luminance difference calculation unit 36, FIG. 23 (a) shows a bird's-eye view image in a bird's-eye view state, and FIG. 23 (b) is shown in FIG. 23 (a). It is the figure which expanded a part B1 of the bird's-eye view image. 23, only the detection area A1 is illustrated and described, but the luminance difference is calculated in the same procedure for the detection area A2.

When the adjacent vehicle V2 is reflected in the captured image captured by the camera 10, the adjacent vehicle V2 appears in the detection area A1 in the bird's-eye view image as shown in FIG. As shown in the enlarged view of the region B1 in FIG. 23A in FIG. 23B, it is assumed that the attention line La is set on the rubber portion of the tire of the adjacent vehicle V2 on the bird's-eye view image. In this state, the luminance difference calculation unit 36 first sets a reference line Lr. The reference line Lr is set along the vertical direction at a position away from the attention line La by a predetermined distance in the real space. Specifically, in the three-dimensional object detection device 1a according to the present embodiment, the reference line Lr is set at a position separated from the attention line La by 10 cm in the real space. Thereby, the reference line Lr is set on the wheel of the tire of the adjacent vehicle V2, which is separated from the rubber of the tire of the adjacent vehicle V2, for example, by 10 cm, on the bird's eye view image.

Next, the luminance difference calculation unit 36 sets a plurality of attention points Pa1 to PaN on the attention line La. In FIG. 23B, for the convenience of explanation, six attention points Pa1 to Pa6 (hereinafter simply referred to as attention point Pai when an arbitrary point is indicated) are set. Note that the number of attention points Pa set on the attention line La may be arbitrary. In the following description, it is assumed that N attention points Pa are set on the attention line La.

Next, the luminance difference calculation unit 36 sets the reference points Pr1 to PrN so as to be the same height as the attention points Pa1 to PaN in the real space. Then, the luminance difference calculation unit 36 calculates the luminance difference between the attention point Pa and the reference point Pr having the same height. Accordingly, the luminance difference calculation unit 36 calculates the luminance difference between the two pixels for each of a plurality of positions (1 to N) along the vertical imaginary line extending in the vertical direction in the real space. The luminance difference calculation unit 36 calculates, for example, a luminance difference between the first attention point Pa1 and the first reference point Pr1, and a luminance difference between the second attention point Pa2 and the second reference point Pr2. Will be calculated. Thereby, the luminance difference calculation unit 36 continuously obtains the luminance difference along the attention line La and the reference line Lr. That is, the luminance difference calculation unit 36 sequentially obtains the luminance difference between the third to Nth attention points Pa3 to PaN and the third to Nth reference points Pr3 to PrN.

The luminance difference calculation unit 36 repeatedly executes the processing such as setting the reference line Lr, setting the attention point Pa and the reference point Pr, and calculating the luminance difference while shifting the attention line La in the detection area A1. That is, the luminance difference calculation unit 36 repeatedly executes the above processing while changing the positions of the attention line La and the reference line Lr by the same distance in the extending direction of the ground line L1 in the real space. For example, the luminance difference calculation unit 36 sets a line that has been the reference line Lr in the previous process as the attention line La, sets the reference line Lr for the attention line La, and sequentially obtains the luminance difference. It will be.

As described above, in the second embodiment, the edge extending in the vertical direction is obtained by calculating the luminance difference from the attention point Pa on the attention line La and the reference point Pr on the reference line Lr that are substantially the same height in the real space. It is possible to clearly detect a luminance difference in the case where there is. Also, in order to compare the brightness of vertical virtual lines extending in the vertical direction in real space, even if the three-dimensional object is stretched according to the height from the road surface by converting to a bird's-eye view image, The detection process is not affected, and the detection accuracy of the three-dimensional object can be improved.

20, the edge line detection unit 37 detects the edge line from the continuous luminance difference calculated by the luminance difference calculation unit 36. For example, in the case illustrated in FIG. 23B, the first attention point Pa1 and the first reference point Pr1 are located in the same tire portion, and thus the luminance difference is small. On the other hand, the second to sixth attention points Pa2 to Pa6 are located in the rubber part of the tire, and the second to sixth reference points Pr2 to Pr6 are located in the wheel part of the tire. Therefore, the luminance difference between the second to sixth attention points Pa2 to Pa6 and the second to sixth reference points Pr2 to Pr6 becomes large. Therefore, the edge line detection unit 37 may detect that an edge line exists between the second to sixth attention points Pa2 to Pa6 and the second to sixth reference points Pr2 to Pr6 having a large luminance difference. it can.

Specifically, when detecting the edge line, the edge line detection unit 37 firstly follows the following Equation 2 to determine the i-th attention point Pai (coordinates (xi, yi)) and the i-th reference point Pri (coordinates ( xi ′, yi ′)) and the i th attention point Pai are attributed.
[Equation 2]
When I (xi, yi)> I (xi ′, yi ′) + t s (xi, yi) = 1
When I (xi, yi) <I (xi ′, yi ′) − t s (xi, yi) = − 1
Otherwise s (xi, yi) = 0

In Equation 2, t represents a predetermined threshold, I (xi, yi) represents the luminance value of the i-th attention point Pai, and I (xi ′, yi ′) represents the luminance value of the i-th reference point Pri. Indicates. According to Equation 2, when the luminance value of the attention point Pai is higher than the luminance value obtained by adding the threshold value t to the reference point Pri, the attribute s (xi, yi) of the attention point Pai is “1”. Become. On the other hand, when the luminance value of the attention point Pai is lower than the luminance value obtained by subtracting the luminance threshold t from the reference point Pri, the attribute s (xi, yi) of the attention point Pai is “−1”. When the luminance value of the attention point Pai and the luminance value of the reference point Pri are in other relationships, the attribute s (xi, yi) of the attention point Pai is “0”.

Next, the edge line detection unit 37 determines whether or not the attention line La is an edge line from the continuity c (xi, yi) of the attribute s along the attention line La based on Equation 3 below.
[Equation 3]
When s (xi, yi) = s (xi + 1, yi + 1) (and excluding 0 = 0),
c (xi, yi) = 1
Other than the above
c (xi, yi) = 0

When the attribute s (xi, yi) of the attention point Pai and the attribute s (xi + 1, yi + 1) of the adjacent attention point Pai + 1 are the same, the continuity c (xi, yi) is “1”. When the attribute s (xi, yi) of the attention point Pai is not the same as the attribute s (xi + 1, yi + 1) of the adjacent attention point Pai + 1, the continuity c (xi, yi) is “0”.

Next, the edge line detection unit 37 obtains the sum for the continuity c of all the attention points Pa on the attention line La. The edge line detection unit 37 normalizes the continuity c by dividing the obtained sum of continuity c by the number N of points of interest Pa. Then, the edge line detection unit 37 determines that the attention line La is an edge line when the normalized value exceeds the threshold θ. The threshold value θ is a value set in advance through experiments or the like.

That is, the edge line detection unit 37 determines whether or not the attention line La is an edge line based on Equation 4 below. Then, the edge line detection unit 37 determines whether or not all the attention lines La drawn on the detection area A1 are edge lines.
[Equation 4]
Σc (xi, yi) / N> θ

As described above, in the second embodiment, the attention point Pa is attributed based on the luminance difference between the attention point Pa on the attention line La and the reference point Pr on the reference line Lr, and the attribute along the attention line La is attributed. Since it is determined whether the attention line La is an edge line based on the continuity c of the image, the boundary between the high luminance area and the low luminance area is detected as an edge line, and an edge in line with a natural human sense Detection can be performed. This effect will be described in detail. FIG. 24 is a diagram illustrating an example of an image for explaining the processing of the edge line detection unit 37. In this image example, a first striped pattern 101 showing a striped pattern in which a high brightness area and a low brightness area are repeated, and a second striped pattern showing a striped pattern in which a low brightness area and a high brightness area are repeated. 102 is an adjacent image. Further, in this image example, a region where the brightness of the first striped pattern 101 is high and a region where the brightness of the second striped pattern 102 is low are adjacent to each other, and a region where the brightness of the first striped pattern 101 is low and the second striped pattern 102. Is adjacent to a region with high brightness. The portion 103 located at the boundary between the first striped pattern 101 and the second striped pattern 102 tends not to be perceived as an edge depending on human senses.

On the other hand, since the low luminance region and the high luminance region are adjacent to each other, if the edge is detected only by the luminance difference, the portion 103 is recognized as an edge. However, the edge line detection unit 37 determines the part 103 as an edge line only when the luminance difference attribute has continuity in addition to the luminance difference in the part 103. An erroneous determination of recognizing a part 103 that is not recognized as an edge line as a sensation as an edge line can be suppressed, and edge detection according to a human sensation can be performed.

20, the three-dimensional object detection unit 33a detects a three-dimensional object based on the amount of edge lines detected by the edge line detection unit 37. As described above, the three-dimensional object detection device 1a according to the present embodiment detects an edge line extending in the vertical direction in real space. The fact that many edge lines extending in the vertical direction are detected means that there is a high possibility that a three-dimensional object exists in the detection areas A1 and A2. For this reason, the three-dimensional object detection unit 33a detects a three-dimensional object based on the amount of edge lines detected by the edge line detection unit 37. Specifically, the three-dimensional object detection unit 33a determines whether the amount of edge lines detected by the edge line detection unit 37 is equal to or greater than a predetermined threshold value β, and the amount of edge lines is determined to be a predetermined threshold value β. In the case described above, the edge line detected by the edge line detection unit 37 is determined to be an edge line of a three-dimensional object, thereby detecting the three-dimensional object based on the edge line as the adjacent vehicle V2.

As described above, the edge line is an aspect of pixel distribution information indicating a predetermined luminance difference, and the “pixel distribution information” in the present embodiment is a three-dimensional object when a captured image is converted into a bird's-eye view image. It can be positioned as information indicating the distribution state of “pixels having a luminance difference equal to or greater than a predetermined threshold” detected along the falling direction. That is, in the bird's-eye view image obtained by the viewpoint conversion unit 31, the three-dimensional object detection unit 33a distributes pixels whose luminance difference is greater than or equal to the threshold value t along the direction in which the three-dimensional object falls when the viewpoint is converted into the bird's-eye view image. When the degree of pixel distribution (the amount of edge lines) in the direction in which the three-dimensional object falls is equal to or greater than a predetermined threshold β, the three-dimensional object is detected based on the pixel distribution information (edge line).

In the second embodiment, the threshold value β for detecting the adjacent vehicle based on the edge line is set by the detection reference setting unit 35a. In the second embodiment, the detection reference setting unit 35a is set according to the rear distance from the camera 10 based on the certainty factor detected by the certainty factor determination unit 34 shown in FIG. 20 as in the first embodiment. One of the third threshold value β ₁ and the fourth threshold value β ₂ set according to the positional relationship between the camera 10 and the light source is used as a threshold value β for detecting an adjacent vehicle based on the edge line. Set as.

That is, the detection criterion setting unit 35a determines whether or not the certainty level determined by the certainty level determination unit 34 is equal to or greater than a predetermined value. The third threshold value β ₁ set according to the distance is set as a threshold value β for detecting the adjacent vehicle based on the edge line. On the other hand, when the certainty factor is equal to or greater than a predetermined value, the camera 10 and the light source the fourth threshold value beta ₂ which is set in accordance with the positional relationship between, is set as the threshold value beta for detecting adjacent vehicle based on the edge lines.

In the second embodiment, the certainty factor determination unit 34 is adjacent to the detected light source based on the number of detected light sources, the position of the light sources, and the temporal change in the state of the light sources, as in the first embodiment. The reliability that is the headlight of the vehicle is comprehensively determined. In this way, by comprehensively determining the number of light sources, the positions of the light sources, and the temporal changes in the state of the light sources, it is possible to appropriately determine whether or not the detected light sources are caused by the headlights of adjacent vehicles. Can be determined. Furthermore, as in the first embodiment, the certainty factor determination unit 34 can be configured to take into account the size of the detected light source when determining the certainty factor.

Here, in the second embodiment, detecting the reference portion setting unit 35a, the third threshold value beta _1, it is set as follows. That is, as in the first embodiment, the detection reference setting unit 35a, as shown in FIG. 25, detects positions within the detection areas A1 and A2 (for example, positions corresponding to the attention line La within the detection areas A1 and A2). The third threshold value β ₁ is set to a higher value as the luminance at () is higher. The detection reference setting unit 35a, as shown in FIG. 26, in the region R1 of the detection areas A1, A2, rather than the region R2, by increasing the third gain threshold beta _1, the detection area A1 , A 2, the third threshold value β ₁ set in accordance with the luminance is changed to a higher value than the region R 2 in the detection areas A 1, A 2. Further, the detection reference setting unit 35a, in the region R1 of the detection area A1, A2, as rear distance from the camera 10 is small (closer to the host vehicle), by increasing the third gain threshold beta ₁ The third threshold value β ₁ set in accordance with the brightness is changed to a higher value as it is closer to the host vehicle.

Similarly, detection reference setting unit 35a, as shown in FIG. 22, in the region R3 of the detection area A1, A2, than the region R2 of the detection area A1, A2, the third threshold value beta ₁ of the gain the by increasing, in the region R3 in the detection area A1, A2, compared to the region R2 of the detection area A1, A2, to change to a higher value of the third threshold value beta _1, which is set according to the brightness. Further, the detection reference setting unit 35a, in the area R3 in the detection areas A1, A2 are more rearward distance from the camera 10 is larger (farther from the vehicle), increasing the third gain threshold beta ₁ in, in the region R3 in the detection area A1, A2 is further away from the vehicle, it is changed to a higher value of the third threshold value beta _1, which is set according to the brightness.

Thereby, also in 2nd Embodiment, similarly to 1st Embodiment, the headlight irradiated from the adjacent vehicle V3 in the scene where the light of the headlight of the adjacent vehicle V3 is irradiated to an adjacent lane at night, for example. Even in the region R2 where the luminance of the light of the headlight is high, or in the regions R1, R3 where the luminance of the light of the headlight irradiated from the adjacent vehicle V3 is low, an image of the light of the headlight is displayed on the adjacent vehicle. Can be effectively prevented from being erroneously detected.

In the second embodiment, detecting the reference portion setting unit 35a is the fourth threshold value beta _2, is set as follows. That is, as in the first embodiment, the detection reference setting unit 35a detects in the detection areas A1 and A2 so as to easily detect the tire wheel that is a characteristic part of the adjacent vehicle, as shown in FIG. As shown in FIG. 28, a third threshold map in which a predetermined threshold value β ₂ ′ is set according to the luminance at the position, and a detection region A1 so that no erroneous detection of adjacent vehicles occurs around the headlight. , A2 and a fourth threshold map in which a predetermined threshold β ₂ ″ is set according to the luminance at the detection position in A2. The detection reference setting unit 35a is the same as in the first embodiment. In addition, as shown in FIG. 29, a control map defining the weight wt of the threshold value β ₂ ″ obtained from the fourth threshold value map is provided.

Then, similarly to the first embodiment, the detection criterion setting unit 35a performs the threshold β ₂ ′ obtained by the third threshold map and the threshold β ₂ ″ obtained by the fourth threshold map according to the following formula 5. Is weighted with the weight wt of the threshold value β ₂ ″ defined in the control map shown in FIG. 29 according to the position of the light source, thereby the fourth threshold value β for detecting the adjacent vehicle based on the edge information. ₂ is calculated. Also in the second embodiment, like the first embodiment, the position of the detected light sources, by matching the reference position P _O of the control map shown in FIG. 29, adjustment line of the control map of FIG. 29 Is called.
[Equation 5]
4th threshold value β ₂ = {(1-wt) · threshold value β ₂ ′} + obtained from the third threshold map + (wt · threshold value β ₂ ″ obtained from the fourth threshold map)

Thus, the detection reference setting unit 35a, the third and threshold beta ₁ of which is set in accordance with the rearward distance from the camera 10, a fourth threshold value which is set according to the positional relationship between the camera 10 and the light source beta ₂ And can be set. Then, the detection criterion setting unit 35a determines whether or not the certainty factor determined by the certainty factor determination unit 34 is equal to or greater than a predetermined value. If the certainty factor is less than the predetermined value, the third threshold value β ₁ is set as the threshold value β for detecting the adjacent vehicle based on the edge line. On the other hand, when the certainty factor is equal to or greater than the predetermined value, the fourth threshold value β ₂ is set based on the edge line. Is set as a threshold value β for detecting.

Further, prior to detecting the three-dimensional object, the three-dimensional object detection unit 33a determines whether or not the edge line detected by the edge line detection unit 37 is correct. The three-dimensional object detection unit 33a determines whether or not the luminance change along the edge line of the bird's-eye view image on the edge line is equal to or greater than a predetermined threshold value tb. When the brightness change of the bird's-eye view image on the edge line is equal to or greater than the threshold value tb, it is determined that the edge line has been detected by erroneous determination. On the other hand, when the luminance change of the bird's-eye view image on the edge line is less than the threshold value tb, it is determined that the edge line is correct. The threshold value tb is a value set in advance by experiments or the like.

FIG. 30 is a diagram illustrating the luminance distribution of the edge line, and FIG. 30A illustrates the edge line and the luminance distribution when the adjacent vehicle V2 as a three-dimensional object is present in the detection area A1, and FIG. Indicates an edge line and a luminance distribution when there is no solid object in the detection area A1.

As shown in FIG. 30 (a), it is assumed that the attention line La set in the tire rubber portion of the adjacent vehicle V2 is determined to be an edge line in the bird's-eye view image. In this case, the luminance change of the bird's-eye view image on the attention line La is gentle. This is because the tire of the adjacent vehicle V2 is extended in the bird's-eye view image by converting the image captured by the camera 10 into a bird's-eye view image. On the other hand, as shown in FIG. 30B, it is assumed that the attention line La set in the white character portion “50” drawn on the road surface in the bird's-eye view image is erroneously determined as an edge line. In this case, the brightness change of the bird's-eye view image on the attention line La has a large undulation. This is because a portion with high brightness in white characters and a portion with low brightness such as a road surface are mixed on the edge line.

Based on the difference in luminance distribution on the attention line La as described above, the three-dimensional object detection unit 33a determines whether or not the edge line is detected by erroneous determination. When the luminance change along the edge line is equal to or greater than the predetermined threshold value tb, the three-dimensional object detection unit 33a has detected the edge line by erroneous determination, and the edge line is caused by the three-dimensional object. Judge that it is not. Thereby, white characters such as “50” on the road surface, weeds on the road shoulder, and the like are determined as edge lines, and the detection accuracy of the three-dimensional object is prevented from being lowered. On the other hand, when the change in luminance along the edge line is less than the predetermined threshold value tb, the three-dimensional object detection unit 33a determines that the edge line is an edge line of the three-dimensional object, and the three-dimensional object exists. Judge.

Specifically, the three-dimensional object detection unit 33a calculates the luminance change of the edge line according to any one of the following

formulas

6 and 7. The luminance change of the edge line corresponds to the evaluation value in the vertical direction in the real space. Equation 6 below evaluates the luminance distribution by the sum of the squares of the differences between the i-th luminance value I (xi, yi) on the attention line La and the adjacent i + 1-th luminance value I (xi + 1, yi + 1). . Also, the following Expression 7 is a luminance distribution based on the sum of absolute values of differences between the i-th luminance value I (xi, yi) on the attention line La and the adjacent i + 1-th luminance value I (xi + 1, yi + 1). To evaluate.
[Equation 6]
Evaluation value in the vertical equivalent direction = Σ [{I (xi, yi) −I (xi + 1, yi + 1)} ² ]
[Equation 7]
Evaluation value in the vertical equivalent direction = Σ | I (xi, yi) −I (xi + 1, yi + 1) |

Not only Formula 7 but also Formula 8 below, the attribute b of the adjacent luminance value is binarized using the threshold value t2, and the binarized attribute b is summed for all the attention points Pa. Also good.
[Equation 8]
Evaluation value in the vertical equivalent direction = Σb (xi, yi)
However, when | I (xi, yi) −I (xi + 1, yi + 1) |> t2,
b (xi, yi) = 1
Other than the above
b (xi, yi) = 0

When the absolute value of the luminance difference between the luminance value of the attention point Pai and the luminance value of the reference point Pri is larger than the threshold value t2, the attribute b (xi, yi) of the attention point Pa (xi, yi) is “1”. Become. If the relationship is other than that, the attribute b (xi, yi) of the attention point Pai is '0'. This threshold value t2 is set in advance by an experiment or the like in order to determine that the attention line La is not on the same three-dimensional object. Then, the three-dimensional object detection unit 33a sums the attributes b for all the attention points Pa on the attention line La and obtains an evaluation value in the vertical equivalent direction, whereby the edge line is caused by the three-dimensional object. It is determined whether or not a three-dimensional object exists.

Next, an adjacent vehicle detection method according to the second embodiment will be described. FIG. 31 is a flowchart showing details of the adjacent vehicle detection method according to the present embodiment. In FIG. 31, for the sake of convenience, processing for the detection area A1 will be described, but the same processing is executed for the detection area A2.

First, in step S301, the camera 10 captures an image of a predetermined area specified by the angle of view a and the attachment position, and the computer 30a acquires image data of the captured image P captured by the camera 10. Next, in step S302, the viewpoint conversion unit 31 performs viewpoint conversion on the acquired image data to generate bird's-eye view image data.

Next, in step S303, the luminance difference calculation unit 36 sets the attention line La on the detection area A1. At this time, the luminance difference calculation unit 36 sets a line corresponding to a line extending in the vertical direction in the real space as the attention line La. Next, in step S304, the luminance difference calculation unit 36 sets a reference line Lr on the detection area A1. At this time, the luminance difference calculation unit 36 sets a reference line Lr that corresponds to a line segment extending in the vertical direction in the real space and is separated from the attention line La by a predetermined distance in the real space.

Next, the luminance difference calculation unit 36 sets a plurality of attention points Pa on the attention line La in step S305. At this time, the luminance difference calculation unit 36 sets a number of attention points Pa that do not cause a problem when the edge is detected by the edge line detection unit 37. In step S306, the luminance difference calculation unit 36 sets the reference point Pr so that the attention point Pa and the reference point Pr are substantially the same height in the real space. Thereby, the attention point Pa and the reference point Pr are arranged in a substantially horizontal direction, and it becomes easy to detect an edge line extending in the vertical direction in the real space.

Next, in step S307, the luminance difference calculation unit 36 calculates the luminance difference between the attention point Pa and the reference point Pr that have the same height in the real space. Next, the edge line detection unit 37 calculates the attribute s of each attention point Pa according to Equation 2 above. Next, in step S308, the edge line detection unit 37 calculates the continuity c of the attribute s of each attention point Pa according to the above mathematical formula 3. Next, in step S309, the edge line detection unit 37 determines whether the value obtained by normalizing the total sum of continuity c is greater than the threshold value θ according to the above formula 4. If it is determined that the normalized value is greater than the threshold θ (step S309 = Yes), the edge line detection unit 37 detects the attention line La as an edge line in step S310. Then, the process proceeds to step S311. If it is determined that the normalized value is not greater than the threshold θ (step S309 = No), the edge line detection unit 37 does not detect the attention line La as an edge line, and the process proceeds to step S311.

In step S311, the computer 30a determines whether or not the processing in steps S303 to S310 has been executed for all the attention lines La that can be set on the detection area A1. When it is determined that the above processing has not been performed for all the attention lines La (step S311 = No), the processing returns to step S303, a new attention line La is set, and the processing up to step S311 is repeated. On the other hand, if it is determined that the above process has been performed for all the attention lines La (step S311 = Yes), the process proceeds to step S312.

In step S312, the three-dimensional object detection unit 33a calculates a luminance change along the edge line for each edge line detected in step S310. The three-dimensional object detection unit 33a calculates the luminance change of the edge line according to any one of the

above formulas

6, 7, and 8. Next, in step S313, the three-dimensional object detection unit 33a excludes edge lines whose luminance change is equal to or greater than a predetermined threshold value tb from among the edge lines. That is, it is determined that an edge line having a large luminance change is not a correct edge line, and the edge line is not used for detecting a three-dimensional object. As described above, this is to prevent characters on the road surface, roadside weeds, and the like included in the detection area A1 from being detected as edge lines. Therefore, the predetermined threshold value tb is a value set based on a luminance change generated by characters on the road surface, weeds on the road shoulder, or the like, which is obtained in advance through experiments or the like. On the other hand, the three-dimensional object detection unit 33a determines an edge line whose luminance change is less than the predetermined threshold value tb among the edge lines as an edge line of the three-dimensional object, and thereby detects a three-dimensional object existing in the adjacent vehicle. .

Next, in step S314, the threshold value β for determining whether or not the three-dimensional object detected in step S313 is an adjacent vehicle is set by the detection reference setting unit 35a. Specifically, the detection reference setting unit 35a detects the light source on the rear side of the host vehicle, as in step S105 of the first embodiment, and the detected light source is the next adjacent vehicle on the rear side of the host vehicle. A certainty factor indicating the certainty of being a headlight is determined. Then, when the detection reference setting unit 35a, confidence is not less than the predetermined value, the third threshold value beta _1, and set as a threshold value beta for detecting adjacent vehicle based on the edge lines, whereas, confidence There If it is less than the predetermined value, the fourth threshold value beta _2, is set as the threshold value beta for detecting adjacent vehicle based on the edge lines.

In step S315, the three-dimensional object detection unit 33a determines whether the amount of the edge line is equal to or larger than the threshold value β set in step S314. When it is determined that the amount of the edge line is equal to or larger than the threshold value β (step S315 = Yes), the three-dimensional object detection unit 33a determines in step S316 that an adjacent vehicle exists in the detection area A1. On the other hand, when it is determined that the amount of the edge line is not equal to or greater than the threshold value β (step S315 = No), the three-dimensional object detection unit 33a determines in step S317 that there is no adjacent vehicle in the detection area A1. Then, the process shown in FIG. 31 is complete | finished.

As described above, in the second embodiment, a captured image is converted into a bird's-eye view image, and edge information of the three-dimensional object is detected from the converted bird's-eye view image. Then, it is determined whether or not the amount of edge lines detected in the bird's-eye view image is equal to or larger than a threshold value β set according to the positional relationship between the camera 10 and the light source, and the amount of edge lines is equal to or larger than the threshold value β. In this case, it is possible to appropriately detect the three-dimensional object existing in the adjacent lane by determining that the three-dimensional object exists in the adjacent lane. Moreover, in 2nd Embodiment, when detecting an adjacent vehicle based on edge information, the light source of the back side of the own vehicle is detected, and the certainty that the detected light source is a headlight of the adjacent vehicle adjacent to the vehicle is detected. judge. When the certainty factor is less than the predetermined value, the third threshold value β ₁ obtained based on the rear distance from the camera 10 is set as the threshold value β, while when the certainty factor is equal to or larger than the predetermined value. Sets the fourth threshold β ₂ obtained based on the positional relationship between the camera 10 and the light source as the threshold β. As a result, in the second embodiment, in addition to the effects of the first embodiment, when detecting the adjacent vehicle based on the edge information, if the detected light source is the headlight of the adjacent vehicle, the dirt on the lens. Even when the brightness around the headlight (light source) is high, the tire wheel of the adjacent vehicle, which is a characteristic part of the adjacent vehicle existing behind the headlight (light source), can be detected appropriately, and the head In front of the light (light source), it is possible to effectively prevent erroneous detection of the adjacent vehicle due to the reflected light of the headlight of the adjacent vehicle adjacent to the road, and even if it is not the headlight of the adjacent vehicle Adjacent vehicles can be detected appropriately.

The embodiment described above is described for easy understanding of the present invention, and is not described for limiting the present invention. Therefore, each element disclosed in the above embodiment is intended to include all design changes and equivalents belonging to the technical scope of the present invention.

For example, in the first embodiment described above, as shown in Equation 1, the threshold value α ₂ ′ obtained from the first threshold map shown in FIG. 10 and the threshold value α obtained from the second threshold map shown in FIG. ₂ ″ is weighted with the weight wt defined in the control map shown in FIG. 12 and the second threshold value α ₂ is calculated. However, the present invention is not limited to this configuration. For example, FIG. Without using the second threshold map shown in FIG. 4, the threshold value α ₂ ′ obtained from the first threshold map shown in FIG. 10 is set to the weight wt defined in the control map shown in FIG. in weighted, it may calculate the second threshold value alpha _2.
[Equation 9]
Second threshold α ₂ = (wt · threshold α ₂ ′ obtained from first threshold map)

In the control map shown in FIG. 32, the weight wt at the reference position P _O has become a x (x> 1), in the rear from the reference position P _O, the weight wt is set to be 1 . Further, in the control map shown in FIG. 32, similarly to the control map shown in FIG. 12, in front of the reference position P _O, than the rear reference position P _O, the weight wt is remained at a high value. Therefore, even in this case, when the position of the light source and the reference position _PO are made to correspond to each other, the adjacent vehicle existing behind the headlight (light source) is effectively prevented while erroneously detecting adjacent vehicles around the headlight. The tire wheel in the lane can be detected appropriately, and as a result, the adjacent vehicle traveling in the adjacent lane can be detected appropriately.

In the first embodiment described above, the difference between the pixel values of the bird's-eye view images PB _t and PB _t−1 is converted into an absolute value, and when the absolute value is equal to or greater than a predetermined threshold th, the difference image PD _t The configuration in which the pixel value is “1” and the pixel value of the difference image PD _t is “0” when the absolute value is less than the predetermined threshold th has been exemplified. It is good also as a structure changed according to this positional relationship.

Specifically, when the certainty factor is less than a predetermined value, the detection reference setting unit 35 detects the detection regions A1, 1, for example, as shown in the left diagram of FIG. In the region R1 and the region R3 in A2, the gain of the threshold th is made larger than that in the region R2 in the detection regions A1 and A2. On the other hand, when the certainty factor is greater than or equal to a predetermined value, As shown in the right figure, among the detection areas A1 and A2, in the area Rf in front of the light source, the gain of the threshold th may be larger than in the area Rr behind the light source. Thereby, for example, when the headlight of the adjacent vehicle is irradiated, the light of the headlight of the adjacent vehicle becomes difficult to be detected as the differential waveform DW _t , and the light of the headlight of the adjacent vehicle is used as the adjacent vehicle. It is possible to effectively prevent erroneous detection.

When the certainty factor is equal to or greater than a predetermined value, for example, as shown in the right diagram of FIG. 13A, in the detection region A1, A2, in the region Rf in front of the light source, the region Rr behind the light source. Alternatively, the gain of the threshold th may be increased. Thereby, in the region Rf in front of the light source, the threshold value th is set to a high value with respect to the luminance, so that it is difficult for the light of the headlight of the adjacent vehicle V3 to be detected as the differential waveform DW _t. Thus, it is possible to effectively prevent erroneous detection of the adjacent vehicle due to, for example, dirt or reflected light obtained by reflecting the light of the headlight of the adjacent vehicle V3 adjacent to the road. In addition to this configuration, the threshold th may be changed in accordance with the luminance at the detection positions in the detection areas A1 and A2.

Furthermore, in the first embodiment described above, the alignment unit 32 detects the pixel value of the difference image PD _t as “0” or “1”, and based on the difference image PD _t , the three-dimensional object detection unit 33 detects the pixel value. The configuration in which the three-dimensional object is detected by counting the pixels having the pixel value “1” of the difference image PD _t as the difference pixel DP is exemplified. However, the configuration is not limited to this configuration. The unit 32 detects the pixel value of the difference image PD _t with the absolute value of the difference between the pixel values of the bird's-eye view images PB _t and PB _t−1 , and the solid object detection unit 33 sets a predetermined difference threshold value. It is good also as a structure which counts the pixel exceeding it as the difference pixel DP.

Further, in the first embodiment described above, the captured image at the current time and the image one hour before are converted into a bird's-eye view, and after the alignment of the converted bird's-eye view is performed, the difference image PD _t is generated and generated. Although the difference image PD _t is evaluated along the falling direction (the falling direction of the three-dimensional object when the captured image is converted into a bird's eye view), the configuration for generating the difference waveform DW _t is exemplified, but the present invention is not limited to this. Alternatively, the following configuration may be used. For example, only the image one hour before is converted into a bird's-eye view, and the converted bird's-eye view is converted into an equivalent to the image captured again, and a difference image PD _t is generated from this image and the image at the current time. collapsing the difference image PD _t direction corresponding to the direction (i.e., direction and changing the direction in the direction on the captured image collapse) may be configured to generate a difference waveform DW _t by evaluating along. That is, the three-dimensional object when the image of the current time and the image of one hour before are aligned, the difference image PD _t is generated from the difference between the two images subjected to the alignment, and the difference image PD _t is converted into a bird's eye view The bird's-eye view does not necessarily have to be clearly generated as long as the evaluation can be performed along the direction in which the user falls.

Furthermore, in 1st Embodiment mentioned above, although the vehicle speed of the own vehicle V1 is judged based on the signal from the speed sensor 20, it is not restricted to this, As a structure which estimates a speed from the several image at different time. Good. In this case, the vehicle speed sensor 20 is not necessary, and the configuration can be simplified.

In addition, in the first embodiment described above, when the certainty factor is equal to or greater than a predetermined value, the threshold value α ₂ ′ obtained from the first threshold map and the first value are used in order to appropriately detect the tire wheel of the adjacent vehicle. Although the threshold value α ₂ ″ obtained from the threshold map of ₂ is weighted with the weight wt defined in the control map shown in FIG. 12, the configuration in which the threshold value α is lowered behind the position of the light source is illustrated. In addition to this configuration, for example, a distance L in the vehicle width direction from the center position of the camera 10 to the gravity center position of the light source is calculated, and a threshold value behind the reference threshold value P _O corresponding to the light source is calculated according to the distance L. For example, as the distance L in the vehicle width direction from the center position of the camera 10 to the center of gravity of the light source increases, the detected light source travels in an adjacent lane farther from the own lane. The headlight of the adjacent vehicle may be configured to increase the threshold value α at the rear of the light source by determining that the headlight of the adjacent vehicle is highly likely. It is possible to effectively prevent erroneous detection of an adjacent vehicle due to the light.

In the second embodiment described above, when the luminance difference between the attention point Pai and the reference point Pri is equal to or greater than the threshold value t, the attribute s of the attention point Pai is set to “1” or “−1”, An attention line La in which the attribute s (xi, yi) of the point Pai and the attribute s (xi + 1, yi + 1) of the adjacent attention point Pai + 1 are “1” or “−1” continuously over a predetermined threshold θ Although the configuration for detecting the line is illustrated, in addition to this configuration, the threshold value t and the threshold value θ may be changed according to the position of the detected light source.

Specifically, when the certainty factor is less than a predetermined value, the detection reference setting unit 35 detects the detection region in the regions R1 and R3 in the detection regions A1 and A2, for example, as shown in the left diagram of FIG. When the gains of the threshold value t and the threshold value θ are made larger than the area R2 in A1 and A2, and the certainty level is equal to or greater than a predetermined value, for example, as shown in the right figure of FIG. Of A1 and A2, in the region Rf in front of the light source, the gains of the threshold value t and the threshold value θ may be larger than those in the region Rr behind the light source. As a result, when the detected light source is the headlight of the adjacent vehicle adjacent to the headlight, it is difficult to detect the light of the headlight of the adjacent vehicle adjacent to the headlight. It is possible to effectively prevent erroneous detection.

In addition, in the second embodiment described above, the luminance change of the edge line is calculated according to any one of the

above formulas

6, 7, and 8, and among the edge lines, the edge line whose luminance change is equal to or greater than a predetermined threshold value tb. However, in addition to this configuration, the threshold value tb may be changed according to the position of the detected light source.

Specifically, when the certainty factor is less than a predetermined value, the detection reference setting unit 35 detects the detection region in the regions R1 and R3 in the detection regions A1 and A2, for example, as shown in the left diagram of FIG. When the gain of the threshold value tb is made larger than the region R2 in A1 and A2, and the certainty factor is greater than or equal to the predetermined value, for example, as shown in the right diagram of FIG. 13A, the detection regions A1 and A2 Among these, in the region Rf in front of the light source, the gain of the threshold value tb may be larger than that in the region Rr behind the light source. Also in this case, when the detected light source is the headlight of the adjacent vehicle, the light of the headlight of the adjacent vehicle becomes difficult to be detected as an edge line, and the headlight light image of the adjacent vehicle is Can be effectively prevented from being erroneously detected.

The camera 10 of the above-described embodiment corresponds to the imaging unit of the present invention, the viewpoint conversion unit 31 corresponds to the image conversion unit of the present invention, and the alignment unit 32, the three-dimensional object detection units 33 and 33a, and the detection reference setting. The

units

35 and 35a, the luminance difference calculation unit 36, and the edge line detection unit 37 correspond to the three-dimensional object detection unit of the present invention, and the detection

reference setting units

35 and 35a correspond to the light source detection unit of the present invention.

DESCRIPTION OF SYMBOLS 1, 1a ... Three-dimensional object detection apparatus 10 ... Camera 20 ...

Vehicle speed sensor

30, 30a ... Computer 31 ... Viewpoint conversion part 32 ... Position alignment part 33, 33a ... Three-dimensional object detection part 34 ...

Certainty determination part

35, 35a ... Detection standard Setting unit 36 ... luminance difference calculation unit 37 ... edge line detection unit a ... view angle A1, A2 ... detection region CP ... intersection DP ... difference pixels DW _t , DW _t '... difference waveforms DW _t1 to DW _m , DW _{m + k} to DW _{tn ...} small region L1, L2 ... ground line La, Lb ... line P ... captured image PB t _... bird's-eye view image PD t _... difference image V1 ... vehicle V2 ... adjacent vehicle V3 ... next adjacent vehicle on the direction in which the three-dimensional object collapses

Claims

Imaging means for imaging a predetermined area based on an adjacent lane behind the host vehicle;
Image conversion means for converting the captured image obtained by the imaging means into a bird's eye view image;
Pixels that align the positions of the bird's-eye view images at different times obtained by the image conversion means on the bird's-eye view, and that show a difference equal to or greater than a predetermined first threshold on the difference image of the aligned bird's-eye view images The differential waveform information is generated by counting the number and performing frequency distribution, and when the differential waveform information is equal to or greater than a predetermined second threshold, the three-dimensional object existing in the adjacent lane is detected based on the differential waveform information Three-dimensional object detection means
Light source detection means for detecting a light source present behind the host vehicle based on a captured image obtained by the imaging means;
When the light source is detected by the light source detection means, in the front area ahead of the line connecting the light source and the imaging means in the predetermined area, behind the line connecting the light source and the imaging means And a control unit that sets the first threshold value or the second threshold value high so that the three-dimensional object is difficult to be detected in the rear region of the three-dimensional object detection device.
The three-dimensional object detection device according to claim 1,
The control means sets the first threshold value and the second threshold value high so that when the light source is detected by the light source detection means, it is difficult to detect the three-dimensional object in the front area of the predetermined area. A three-dimensional object detection device.
Imaging means for imaging a predetermined detection target area including an adjacent lane behind the host vehicle;
Image conversion means for converting the image obtained by the imaging means into a bird's eye view image;
When edge information greater than or equal to a predetermined first threshold is detected from a bird's eye view image obtained by the image conversion means, and the edge information is greater than or equal to a predetermined second threshold, it exists in an adjacent lane based on the edge information A three-dimensional object detecting means for detecting a three-dimensional object to be performed;
Light source detection means for detecting a light source present behind the host vehicle based on the image obtained by the imaging means;
When the light source is detected by the light source detection means, in the front area ahead of the line connecting the light source and the imaging means in the predetermined area, behind the line connecting the light source and the imaging means And a control unit that sets the first threshold value or the second threshold value high so that the three-dimensional object is difficult to be detected in the rear region of the three-dimensional object detection device.
The three-dimensional object detection device according to claim 3,
The control means sets the first threshold value and the second threshold value high so that when the light source is detected by the light source detection means, it is difficult to detect the three-dimensional object in the front area of the predetermined area. A three-dimensional object detection device.
The three-dimensional object detection device according to any one of claims 1 to 4,
The control means indicates a first control map indicating a relationship between the first threshold value and the luminance, and indicates a relationship between the first threshold value and the luminance, and the first threshold value is set higher than the first control map. The second control map is used, and is used when detecting the three-dimensional object by weighting the first threshold obtained by the first control map and the first threshold obtained by the second control map. Determining the final first threshold;
The control means increases the weight of the first threshold obtained in the second control map in the forward area of the predetermined area, than the first threshold obtained in the first control map. The three-dimensional object detection device characterized in that in the rear region of the region, the weight of the first threshold value obtained with the first control map is larger than the first threshold value obtained with the second control map.
The three-dimensional object detection device according to any one of claims 1 to 4,
The control means indicates a first control map indicating a relationship between the second threshold and luminance, and indicates a relationship between the second threshold and luminance, and the second threshold is set higher than the first control map. The second control map is used, and is used when detecting the three-dimensional object by weighting the second threshold obtained by the first control map and the second threshold obtained by the second control map. Determining the final second threshold;
The control means increases the weight of the second threshold value obtained by the second control map in the forward region of the predetermined region, than the second threshold value obtained by the first control map, and The three-dimensional object detection device characterized in that, in the rear region of the region, the weight of the second threshold value obtained by the first control map is made larger than the second threshold value obtained by the second control map.
Imaging means for imaging a predetermined area based on an adjacent lane behind the host vehicle;
Image conversion means for converting the captured image obtained by the imaging means into a bird's eye view image;
On the bird's-eye view image obtained by the image conversion means, pixel distribution information having a luminance difference equal to or greater than a predetermined first threshold is detected in a direction in which the three-dimensional object collapses when the viewpoint is converted into the bird's-eye view image. Three-dimensional object detection means for detecting the three-dimensional object based on the distribution information of the pixel when the degree of distribution of the pixel in a direction in which the three-dimensional object falls is equal to or greater than a predetermined second threshold;
A light source detection means for detecting a light source existing behind the host vehicle based on a captured image obtained by the imaging means;
When the light source is detected by the light source detection means, in the front area ahead of the line connecting the light source and the imaging means in the predetermined area, behind the line connecting the light source and the imaging means And a control unit that sets the first threshold value or the second threshold value high so that the three-dimensional object is difficult to be detected in the rear region of the three-dimensional object detection device.